Efficient strategies for leave-one-out cross validation for genomic best linear unbiased prediction

The impact of cannabis use on erectile dysfunction and sex hormones: a Mendelian randomization analysis

Previous study has highlighted an association between cannabis use (CU) and an increased risk of erectile dysfunction (ED), potentially due to indirect effects on sex hormonal balance. However, the evidence remains controversial, and the causal relationship is unclear. This study utilized genome-wide association study (GWAS) data to investigate the causal relationships between cannabis use disorder (CUD), lifetime cannabis use (LCU), and ED, as well as levels of sex hormones including estradiol (E2), bioavailable testosterone (BT), follicle-stimulating hormone (FSH), and luteinizing hormone (LH) through Mendelian randomization (MR) analysis. The primary method of analysis was the inverse variance weighted (IVW) method. Data from the FinnGen and UK Biobank were used for replication and meta-analysis. The results indicated no causal relationship between genetically predicted CUD (OR = 0.97, 95% CI 0.87-1.10, P = 0.66) and LCU (OR = 1.13, 95% CI 0.84-1.50, P = 0.42) with the risk of ED. The meta-analysis provided consistent evidence (P > 0.05). No causal relationships were found between CUD and LCU with E2(CUD: β = 0.00, 95% CI 0.00-0.01, P = 0.37; LCU: β = 0.00, 95% CI -0.02-0.01, P = 0.62), BT (CUD: β = 0.00, 95% CI -0.03-0.02, P = 0.90; LCU: β = 0.02, 95% CI -0.04-0.09, P = 0.46), FSH (CUD: β = 0.01, 95% CI -0.18-0.20, P = 0.92; LCU: β = 0.01, 95% CI -0.44-0.47, P = 0.95), and LH (CUD: β = 0.01, 95% CI -0.18-0.21, P = 0.90; LCU: β = 0.13, 95% CI -0.22-0.49, P = 0.46). Sensitivity analyses detected no evidence of horizontal pleiotropy or heterogeneity, ensuring the robustness of the results. In conclusion, this MR analysis did not provide evidence supporting a causal relationship between CU and ED or sex hormone levels.

Genetic association of serum lipids and lipid-modifying targets with endometriosis: Trans-ethnic Mendelian-randomization and mediation analysis

BACKGROUND

Prior observational research identified dyslipidemia as a risk factor for endometriosis (EMS) but the causal relationship remains unestablished due to inherent study limitations.

METHODS

Genome-wide association study data for high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), triglycerides (TG), and total cholesterol (TC) from European (EUR) and East Asian (EAS) ancestries were sourced from the Global Lipids Genetics Consortium. Multi-ancestry EMS data came from various datasets. Univariable Mendelian randomization (MR) examined causal links between serum lipids and EMS. Multivariable and mediation MR explored the influence of seven confounding factors and mediators. Drug-target MR investigates the association between lipid-lowering target genes identified in positive results and EMS. The primary method was inverse-variance weighted (IVW), with replication datasets and meta-analyses reinforcing causal associations. Sensitivity analyses included false discovery rate (FDR) correction, causal analysis using summary effect estimates (CAUSE), and colocalization analysis.

RESULTS

IVW analysis in EUR ancestry showed a significant causal association between TG and increased EMS risk (OR = 1.112, 95% CI 1.033-1.198, P = 5.03×10-3, PFDR = 0.03), supported by replication and meta-analyses. CAUSE analysis confirmed unbiased results (P < 0.05). Multivariable and mediation MR revealed that systolic blood pressure (Mediation effect: 7.52%, P = 0.02) and total testosterone (Mediation effect: 10.79%, P = 0.01) partly mediated this relationship. No causal links were found between other lipid traits and EMS (P > 0.05 & PFDR > 0.05). In EAS ancestry, no causal relationships with EMS were detected (P > 0.05 & PFDR > 0.05). Drug-target MR indicated suggestive evidence for the influence of ANGPTL3 on EMS mediated through TG (OR = 0.798, 95% CI 0.670-0.951, P = 0.01, PFDR = 0.04, PP.H4 = 0.85%).

CONCLUSIONS

This MR study in EUR ancestry indicated an increased EMS risk with higher serum TG levels.

Thyroid cancer and cardiovascular diseases: a Mendelian randomization study

Background

Multiple observational studies have shown associations between thyroid cancer (TC) and cardiovascular diseases (CVDs). However, the results were inconsistent, and the potential causal genetic relationship remains unclear.

Methods

The genetic instruments of TC and CVDs were derived from data obtained through genome-wide association studies (GWAS). We performed the two-sample Mendelian randomization(MR) methods to investigate the causality of TC on CVDs. Summary-level statistics for CVDs, including heart failure (HF), atrial fibrillation (AF), coronary artery disease (CAD), myocardial infarction (MI), ischemic stroke (IS) and venous thromboembolism (VTE). The primary method employed in this MR analysis was the Inverse Variance Weighted (IVW) approach, and four additional algorithms were used: MR-Egger, weighted median, simple mode, and weighted mode. Additionally, we assessed the reliability of the causal relationship through pleiotropy, heterogeneity and leave-one-out sensitivity analysis.

Results

In this MR analysis, we only detected causality of genetically predicted TC on HF (IVW method, odds ratio (OR) = 1.00134, 95% confidence interval (CI): 1.00023-1.00244, p = 0.017). However, There were no causal associations of TC with CAD, MI, AF, IS, and VTE.

Conclusion

Our results confirmed the causal association between TC and HF. It is crucial to closely monitor the incidence of HF in TC patients and give comprehensive clinical intervention based on conventional treatment.

Autoimmune diseases and their genetic link to bronchiectasis: insights from a genetic correlation and Mendelian randomization study

Background

Previous studies have demonstrated that autoimmune diseases are closely associated with bronchiectasis (BE). However, the causal effects between autoimmune diseases and BE remain elusive.

Methods

All summary-level data were obtained from large-scale Genome-Wide Association Studies (GWAS). The univariate Mendelian randomization (UVMR) was utilized to investigate the genetic causal correlation (rg) of 12 autoimmune diseases and bronchiectasis, The Multivariable Mendelian Randomization (MVMR) method was used to explore the effects of the confounding factors. Further investigation was conducted to identify potential intermediate factors using mediation analysis. Finally, the linkage disequilibrium score regression (LDSC) method was used to identify genetic correlations among complex traits. A series of sensitivity analyses was performed to validate the robustness of the results.

Results

The LDSC analysis revealed significant genetic correlations between BE and Crohn's disease (CD) (rg = 0.220, P = 0.037), rheumatoid arthritis (RA) (rg = 0.210, P = 0.021), and ulcerative colitis (UC) (rg = 0.247, P = 0.023). However, no genetic correlation was found with other autoimmune diseases (P > 0.05). The results of the primary IVW analysis suggested that for every SD increase in RA, there was a 10.3% increase in the incidence of BE (odds ratio [OR] = 1.103, 95% confidence interval [CI] 1.055-1.154, P = 1.75×10-5, FDR = 5.25×10-5). Furthermore, for every standard deviation (SD) increase in celiac disease (CeD), the incidence of BE reduced by 5.1% (OR = 0.949, 95% CI 0.902-0.999, P = 0.044, FDR = 0.044). We also observed suggestive evidence corresponding to a 3% increase in BE incidence with T1DM (OR = 1.033, 95% CI 1.001-1.066, P = 0.042, FDR = 0.063). Furthermore, MVMR analysis showed that RA was an independent risk factor for BE, whereas mediator MR analysis did not identify any mediating factors. The sensitivity analyses corroborated the robustness of these findings.

Conclusion

LDSC analysis revealed significant genetic correlations between several autoimmune diseases and BE, and further MVMR analysis showed that RA is an independent risk factor for BE.

Molecular characterization of chronic cutaneous wounds reveals subregion- and wound type-specific differential gene expression

A limited understanding of the pathology underlying chronic wounds has hindered the development of effective diagnostic markers and pharmaceutical interventions. This study aimed to elucidate the molecular composition of various common chronic ulcer types to facilitate drug discovery strategies. We conducted a comprehensive analysis of leg ulcers (LUs), encompassing venous and arterial ulcers, foot ulcers (FUs), pressure ulcers (PUs), and compared them with surgical wound healing complications (WHCs). To explore the pathophysiological mechanisms and identify similarities or differences within wounds, we dissected wounds into distinct subregions, including the wound bed, border, and peri-wound areas, and compared them against intact skin. By correlating histopathology, RNA sequencing (RNA-Seq), and immunohistochemistry (IHC), we identified unique genes, pathways, and cell type abundance patterns in each wound type and subregion. These correlations aim to aid clinicians in selecting targeted treatment options and informing the design of future preclinical and clinical studies in wound healing. Notably, specific genes, such as PITX1 and UPP1, exhibited exclusive upregulation in LUs and FUs, potentially offering significant benefits to specialists in limb preservation and clinical treatment decisions. In contrast, comparisons between different wound subregions, regardless of wound type, revealed distinct expression profiles. The pleiotropic chemokine-like ligand GPR15L (C10orf99) and transmembrane serine proteases TMPRSS11A/D were significantly upregulated in wound border subregions. Interestingly, WHCs exhibited a nearly identical transcriptome to PUs, indicating clinical relevance. Histological examination revealed blood vessel occlusions with impaired angiogenesis in chronic wounds, alongside elevated expression of genes and immunoreactive markers related to blood vessel and lymphatic epithelial cells in wound bed subregions. Additionally, inflammatory and epithelial markers indicated heightened inflammatory responses in wound bed and border subregions and reduced wound bed epithelialization. In summary, chronic wounds from diverse anatomical sites share common aspects of wound pathophysiology but also exhibit distinct molecular differences. These unique molecular characteristics present promising opportunities for drug discovery and treatment, particularly for patients suffering from chronic wounds. The identified diagnostic markers hold the potential to enhance preclinical and clinical trials in the field of wound healing.

Genomic selection in plant breeding: Key factors shaping two decades of progress

Genomic selection, the application of genomic prediction (GP) models to select candidate individuals, has significantly advanced in the past two decades, effectively accelerating genetic gains in plant breeding. This article provides a holistic overview of key factors that have influenced GP in plant breeding during this period. We delved into the pivotal roles of training population size and genetic diversity, and their relationship with the breeding population, in determining GP accuracy. Special emphasis was placed on optimizing training population size. We explored its benefits and the associated diminishing returns beyond an optimum size. This was done while considering the balance between resource allocation and maximizing prediction accuracy through current optimization algorithms. The density and distribution of single-nucleotide polymorphisms, level of linkage disequilibrium, genetic complexity, trait heritability, statistical machine-learning methods, and non-additive effects are the other vital factors. Using wheat, maize, and potato as examples, we summarize the effect of these factors on the accuracy of GP for various traits. The search for high accuracy in GP-theoretically reaching one when using the Pearson's correlation as a metric-is an active research area as yet far from optimal for various traits. We hypothesize that with ultra-high sizes of genotypic and phenotypic datasets, effective training population optimization methods and support from other omics approaches (transcriptomics, metabolomics and proteomics) coupled with deep-learning algorithms could overcome the boundaries of current limitations to achieve the highest possible prediction accuracy, making genomic selection an effective tool in plant breeding.

Causal relationship between cigarette smoking behaviors and the risk of hernias: a Mendelian randomization study

PURPOSE

As the global population continues to age, there is a noticeable yearly rise in the incidence of hernias. Simultaneously, smoking, a widespread addictive behavior and a significant contributor to mortality, has evolved into a pervasive public health concern. Existing literature has already established a connection between smoking and an increased risk of postoperative recurrence and postoperative infections following hernia surgery. However, there remains a dearth of research exploring the association between smoking and hernia morbidity. In this study, our objective is to systematically evaluate the causal relationship between cigarette smoking behaviors and hernia morbidity using a Mendelian randomization (MR) approach.

METHODS

Hernia-related data were sourced from the FinnGen Biobank database, while cigarette smoking behavior data were gathered from the GWAS and Sequencing Consortium of Alcohol and Nicotine Use. To assess the causal relationship, we employed five methods: the weighted median, the weighted mode the inverse variance weighted (IVW), MR-Egger, and the simple mode. Sensitivity analysis was conducted, incorporating Cochran's Q test, the MR-Egger intercept test, leave-one-out analysis, and funnel plot. The presentation of the causal relationship is expressed as an odds ratio (OR) along with their corresponding 95% confidence intervals (CI).

RESULTS

Employing the IVW method as the reference standard, we found that smoking intensity is associated with an increased risk of diaphragmatic hernia (OR = 1.21, 95% CI 1.00-1.46, P = 0.047). These consistent findings were further corroborated by the weighted median and weighted mode methods (OR = 1.26, 95% CI 1.03-1.54, P = 0.026; OR = 1.25, 95% CI 1.02-1.52, P = 0.045). Conversely, when applying the IVW method, we identified no statistically significant causal relationship between smoking age, smoking initiation status, smoking cessation status, and the incidence of hernia.

CONCLUSIONS

Our MR study has uncovered genetic evidence linking smoking intensity and the occurrence of diaphragmatic hernia. The risk of developing diaphragmatic hernia rises in tandem with the intensity of smoking. This emphasizes the crucial role of regularly advising patients to cease smoking in clinical settings.

The protective role of oily fish intake against type 2 diabetes: insights from a genetic correlation and Mendelian randomization study

Background and aims

Previous research has underscored the association between oily fish intake and type 2 diabetes (T2DM), yet the causality remains elusive.

Methods

A bidirectional univariable Mendelian Randomization (MR) analysis was employed to evaluate the causal effects of oily fish and non-oily fish intake on T2DM. Replication analysis and meta-analysis were conducted to ensure robust results. Multivariable MR analysis was utilized to assess confounders, and further mediation MR analysis discerned mediating effects. Linkage Disequilibrium Score (LDSC) analysis was undertaken to compute genetic correlations. Inverse variance weighted (IVW) was the primary method, complemented by a series of sensitivity analyses.

Results

The LDSC analysis unveiled a significant genetic correlation between oily fish intake and T2DM (Genetic correlation: -0.102, p = 4.43 × 10-4). For each standard deviation (SD) increase in genetically predicted oily fish intake, the risk of T2DM was reduced by 38.6% (OR = 0.614, 95% CI 0.504 ~ 0.748, p = 1.24 × 10-6, False Discovery Rate (FDR) = 3.72 × 10-6). The meta-analysis across three data sources highlighted a persistent causal association (OR = 0.728, 95% CI 0.593 ~ 0.895, p = 0.003). No other causal effects were identified (all p > 0.5, FDR > 0.5). The main outcomes remained consistent in most sensitivity analyses. Both MVMR and mediation MR analyses emphasized the mediating roles of triglycerides (TG), body mass index (BMI), and 25-hydroxyvitamin D (25OHD) levels.

Conclusion

To encapsulate, there's an inverse association between oily fish intake and T2DM risk, suggesting potential benefits of oily fish intake in T2DM prevention.

Causal effects between gut microbiota and pulmonary arterial hypertension: A bidirectional Mendelian randomization study

BACKGROUND

Multiple studies have highlighted a potential link between gut microbes and the onset of Pulmonary Arterial Hypertension (PAH). Nonetheless, the precise cause-and-effect relationship remains uncertain.

OBJECTIVES

In this investigation, we utilized a two-sample Mendelian randomization (TSMR) approach to probe the presence of a causal connection between gut microbiota and PAH.

METHODS

Genome-wide association (GWAS) data for gut microbiota and PAH were sourced from MiBioGen and FinnGen research, respectively. Inverse variance weighting (IVW) was used as the primary method to explore the causal effect between gut flora and PAH, supplemented by MR-Egger, weighted median (WM). Sensitivity analyses examined the robustness of the MR results. Reverse MR analysis was used to rule out the effect of reverse causality on the results.

RESULTS

The results indicate that Genus Ruminococcaceae UCG004 (OR = 0.407, P = 0.031) and Family Alcaligenaceae (OR = 0.244, P = 0.014) were protective factors for PAH. Meanwhile Genus Lactobacillus (OR = 2.446, P = 0.013), Class Melainabacteria (OR = 2.061, P = 0.034), Phylum Actinobacteria (OR = 3.406, P = 0.010), Genus Victivallis (OR = 1.980, P = 0.010), Genus Dorea (OR = 3.834, P = 0.024) and Genus Slackia (OR = 2.622, P = 0.039) were associated with an increased Prevalence of PAH. Heterogeneity and pleiotropy were not detected by sensitivity analyses, while there was no reverse causality for these nine specific gut microorganisms.

CONCLUSIONS

This study explores the causal effects of eight gut microbial taxa on PAH and provides new ideas for early prevention of PAH.

Design of new molecules against cervical cancer using DFT, theoretical spectroscopy, 2D/3D-QSAR, molecular docking, pharmacophore and ADMET investigations

Cervical cancer is a major health problem of women. Hormone therapy, via aromatase inhibition, has been proposed as a promising way of blocking estrogen production as well as treating the progression of estrogen-dependent cancer. To overcome the challenging complexities of costly drug design, in-silico strategy, integrating Structure-Based Drug Design (SBDD) and Ligand-Based Drug Design (LBDD), was applied to large representative databases of 39 quinazoline and thioquinazolinone compound derivatives. Quantum chemical and physicochemical descriptors have been investigated using density functional theory (DFT) and MM2 force fields, respectively, to develop 2D-QSAR models, while CoMSIA and CoMFA descriptors were used to build 3D-QSAR models. The robustness and predictive power of the reliable models were verified, via several validation methods, leading to the design of 6 new drug-candidates. Afterwards, 2 ligands were carefully selected using virtual screening methods, taking into account the applicability domain, synthetic accessibility, and Lipinski's criteria. Molecular docking and pharmacophore modelling studies were performed to examine potential interactions with aromatase (PDB ID: 3EQM). Finally, the ADMET properties were investigated in order to select potential drug-candidates against cervical cancer for experimental in vitro and in vivo testing.

The causal correlation between gut microbiota abundance and pathogenesis of cervical cancer: a bidirectional mendelian randomization study

Background

Observational studies and animal experiments suggested potential relevance between gut microbiota (GM) and cervical cancer (CC), but the relevance of this association remains to be clarified.

Methods

We performed a two-sample bidirectional Mendelian randomization (MR) analysis to explore whether there was a causal correlation between GM and CC, and the direction of causality.

Results

In primary outcomes, we found that a higher abundance of class Clostridia, family Family XI, genus Alloprevotella, genus Ruminiclostridium 9, and order Clostridiales predicted higher risk of CC, and a higher abundance of class Lentisphaeria, family Acidaminococcaceae, genus Christensenellaceae R7 group, genus Marvinbryantia, order Victivallales, phylum Actinobacteria, and phylum Lentisphaerae predicted lower risk of CC. During verifiable outcomes, we found that a higher abundance of class Methanobacteria, family Actinomycetaceae, family Methanobacteriaceae, genus Lachnospiraceae UCG 010, genus Methanobrevibacter, order Actinomycetales, and order Methanobacteriales predicted a higher risk of CC, and a higher abundance of family Streptococcaceae, genus Dialister, and phylum Bacteroidetes predicted a lower risk of CC, and vice versa.

Conclusion

Our study implied a mutual causality between GM and CC, which provided a novel concept for the occurrence and development of CC, and might promote future functional or clinical analysis.

Associations between artificial sweetener intake from cereals, coffee, and tea and the risk of type 2 diabetes mellitus: A genetic correlation, mediation, and mendelian randomization analysis

BACKGROUND

Previous studies have emphasized the association between the intake of artificial sweeteners (AS) and type 2 diabetes mellitus (T2DM), but the causative relationship remains ambiguous.

METHODS

This study employed univariate Mendelian randomization (MR) analysis to assess the causal link between AS intake from various sources and T2DM. Linkage disequilibrium score (LDSC) regression was used to evaluate the correlation between phenotypes. Multivariate and mediation MR were applied to investigate confounding factors and mediating effects. Data on AS intake from different sources (N = 64,949) were sourced from the UK Biobank, while T2DM data were derived from the DIAbetes Genetics Replication And Meta-analysis.The primary method adopted was inverse variance weighted (IVW), complemented by three validation techniques. Additionally, a series of sensitivity analyses were performed to evaluate pleiotropy and heterogeneity.

RESULTS

LDSC analysis unveiled a significant genetic correlation between AS intake from different sources and T2DM (rg range: -0.006 to 0.15, all P < 0.05). After correction by the false discovery rate (FDR), the primary IVW method indicated that AS intake in coffee was a risk factor for T2DM (OR = 1.265, 95% CI: 1.035-1.545, P = 0.021, PFDR = 0.042). Further multivariable and mediation MR analyses pinpointed high density lipoprotein-cholesterol (HDL-C) as mediating a portion of this causal relationship. In reverse MR analysis, significant evidence suggested a positive correlation between T2DM and AS intake in coffee (β = 0.013, 95% CI: 0.004-0.022, P = 0.004, PFDR = 0.012), cereal (β = 0.007, 95% CI: 0.002-0.012, P = 0.004, PFDR = 0.012), and tea (β = 0.009, 95% CI: 0.001-0.017, P = 0.036, PFDR = 0.049). No other causal associations were identified (P > 0.05, PFDR > 0.05).

CONCLUSION

The MR analysis has established a causal relationship between AS intake in coffee and T2DM. The mediation by HDL-C emphasizes potential metabolic pathways underpinning these relationships.

Digital image analysis and machine learning-assisted prediction of neoadjuvant chemotherapy response in triple-negative breast cancer

BACKGROUND

Pathological complete response (pCR) is associated with favorable prognosis in patients with triple-negative breast cancer (TNBC). However, only 30-40% of TNBC patients treated with neoadjuvant chemotherapy (NAC) show pCR, while the remaining 60-70% show residual disease (RD). The role of the tumor microenvironment in NAC response in patients with TNBC remains unclear. In this study, we developed a machine learning-based two-step pipeline to distinguish between various histological components in hematoxylin and eosin (H&E)-stained whole slide images (WSIs) of TNBC tissue biopsies and to identify histological features that can predict NAC response.

METHODS

H&E-stained WSIs of treatment-naïve biopsies from 85 patients (51 with pCR and 34 with RD) of the model development cohort and 79 patients (41 with pCR and 38 with RD) of the validation cohort were separated through a stratified eightfold cross-validation strategy for the first step and leave-one-out cross-validation strategy for the second step. A tile-level histology label prediction pipeline and four machine-learning classifiers were used to analyze 468,043 tiles of WSIs. The best-trained classifier used 55 texture features from each tile to produce a probability profile during testing. The predicted histology classes were used to generate a histology classification map of the spatial distributions of different tissue regions. A patient-level NAC response prediction pipeline was trained with features derived from paired histology classification maps. The top graph-based features capturing the relevant spatial information across the different histological classes were provided to the radial basis function kernel support vector machine (rbfSVM) classifier for NAC treatment response prediction.

RESULTS

The tile-level prediction pipeline achieved 86.72% accuracy for histology class classification, while the patient-level pipeline achieved 83.53% NAC response (pCR vs. RD) prediction accuracy of the model development cohort. The model was validated with an independent cohort with tile histology validation accuracy of 83.59% and NAC prediction accuracy of 81.01%. The histological class pairs with the strongest NAC response predictive ability were tumor and tumor tumor-infiltrating lymphocytes for pCR and microvessel density and polyploid giant cancer cells for RD.

CONCLUSION

Our machine learning pipeline can robustly identify clinically relevant histological classes that predict NAC response in TNBC patients and may help guide patient selection for NAC treatment.

Genetic Insights into the causal relationship between cannabis use and diabetic phenotypes: A genetic correlation and Mendelian randomization study

BACKGROUND

Previous studies have highlighted the association between cannabis use and diabetes and its complications; however, the causality remains ambiguous.

METHODS

Univariate Mendelian randomization (MR), multivariate MR, mediation MR, and linkage disequilibrium score (LDSC) analysis to assess the causal relationship between cannabis use and 12 diabetic phenotypes. Summary statistics for lifetime cannabis use (N = 184,765) and cannabis use disorder (CUD) (N = 374,287) from genome-wide association studies. The primary method used was inverse-variance-weighted (IVW). A range of sensitivity analyses ensured the robustness of the results.

RESULTS

LDSC analysis revealed a significant genetic correlation between CUD and T2DM, as well as between lifetime cannabis use and four diabetic phenotypes (P < 0.05). After correction by false discovery rate (FDR), the primary IVW analysis indicates that the genetically predicted CUD is positively associated with the risk of diabetic hypoglycemia (OR = 1.11, 95% CI 1.04-1.20, P = 0.003, PFDR = 0.04) and proliferative diabetic retinopathy (PDR) (OR = 1.12, 95% CI 1.04-1.19, P = 4.89×10-4, PFDR =0.01). Additionally, suggestive evidence links CUD with increased risks of diabetic nephropathy, type 1 diabetes mellitus (T1DM), diabetic retinopathy, and T1DM associated with diabetic ketoacidosis (P < 0.05 & PFDR > 0.05). No causal relationship was detected between lifetime cannabis use and diabetic phenotypes (P > 0.05 & PFDR > 0.05). Multivariable and mediation MR analyses revealed that glycated hemoglobin A1c partially mediates the causal effect of CUD on PDR and diabetic hypoglycemia.

CONCLUSION

This MR study suggests that CUD may have a causal role in several diabetic disease phenotypes.

Vitamin D and hyperemesis gravidarum: A mendelian randomization study

BACKGROUND

The causality between vitamin D and hyperemesis gravidarum remains unknown. Our aim was to investigate the causal effect of vitamin D on hyperemesis gravidarum using the two-sample Mendelian randomization method.

METHODS

Independent single nucleotide polymorphisms significantly associated with serum 25-hydroxyvitamin D levels served as instrumental variables. The corresponding effect estimates for hyperemesis gravidarum were obtained from the Finngen Biobank. For Mendelian randomization analysis, inverse variance weighting was used as the primary method. We also used weighted median, MR-Egger regression, simple mode, and weighted mode as complementary methods to inverse variance weighting. The MR-Egger intercept test, Cochran's Q test, and "leave-one-out" sensitivity analysis were performed to assess the horizontal pleiotropy, heterogeneity, and stability of the causal association between 25-hydroxyvitamin D levels and hyperemesis gravidarum.

RESULTS

We found that an increase in 25-hydroxyvitamin D level was associated with a lower risk of hyperemesis gravidarum [odds ratio (OR): 0.568, 95 % CI: 0.403-0.800, p = 0.001]. The result demonstrates the causal relationship between 25-hydroxyvitamin D level and the risk of hyperemesis gravidarum in the European population.

CONCLUSIONS

The large Mendelian randomization analysis suggests that vitamin D may be causally associated with risk of hyperemesis gravidarum.

Using clinical and radiomic feature-based machine learning models to predict pathological complete response in patients with esophageal squamous cell carcinoma receiving neoadjuvant chemoradiation

OBJECTIVE

This study aimed to build radiomic feature-based machine learning models to predict pathological clinical response (pCR) of neoadjuvant chemoradiation therapy (nCRT) for esophageal squamous cell carcinoma (ESCC) patients.

METHODS

A total of 112 ESCC patients who underwent nCRT followed by surgical treatment from January 2008 to December 2018 were recruited. According to pCR status (no visible cancer cells in primary cancer lesion), patients were categorized into primary cancer lesion pCR (ppCR) group (N = 65) and non-ppCR group (N = 47). Patients were also categorized into total pCR (tpCR) group (N = 48) and non-tpCR group (N = 64) according to tpCR status (no visible cancer cells in primary cancer lesion or lymph nodes). Radiomic features of pretreatment CT images were extracted, feature selection was performed, machine learning models were trained to predict ppCR and tpCR, respectively.

RESULTS

A total of 620 radiomic features were extracted. For ppCR prediction models, radiomic model had an area under the curve (AUC) of 0.817 (95% CI: 0.732-0.896) in the testing set; and the combination model that included rad-score and clinical features had a great predicting performance, with an AUC of 0.891 (95% CI: 0.823-0.950) in the testing set. For tpCR prediction models, radiomic model had an AUC of 0.713 (95% CI: 0.613-0.808) in the testing set; and the combination model also had a great predicting performance, with an AUC of 0.814 (95% CI: 0.728-0.881) in the testing set.

CONCLUSION

This study built machine learning models for predicting ppCR and tpCR of ESCC patients with favorable predicting performance respectively, which aided treatment plan optimization.

CLINICAL RELEVANCE STATEMENT

This study significantly improved the predictive value of machine learning models based on radiomic features to accurately predict response to therapy of esophageal squamous cell carcinoma patients after neoadjuvant chemoradiation therapy, providing guidance for further treatment.

KEY POINTS

• Combination model that included rad-score and clinical features had a great predicting performance. • Primary tumor pCR predicting models exhibit better predicting performance compared to corresponding total pCR predicting models.

Serum uric acid and risk of diabetic neuropathy: a genetic correlation and mendelian randomization study

Background

Previous observational studies have indicated an association between serum uric acid (SUA) and diabetic neuropathy (DN), but confounding factors and reverse causality have left the causality of this relationship uncertain.

Methods

Univariate Mendelian randomization (MR), multivariate MR and linkage disequilibrium score (LDSC) regression analysis were utilized to assess the causal link between SUA and DN. Summary-level data for SUA were drawn from the CKDGen consortium, comprising 288,648 individuals, while DN data were obtained from the FinnGen consortium, with 2,843 cases and 271,817 controls. Causal effects were estimated primarily using inverse variance weighted (IVW) analysis, supplemented by four validation methods, with additional sensitivity analyses to evaluate pleiotropy, heterogeneity, and result robustness.

Results

The LDSC analysis revealed a significant genetic correlation between SUA and DN (genetic correlation = 0.293, P = 2.60 × 10-5). The primary methodology IVW indicated that each increase of 1 mg/dL in SUA would increase DN risk by 17% (OR = 1.17, 95% CI 1.02-1.34, P = 0.02), while no causal relationship was found in reverse analysis (OR = 1.00, 95% CI 0.98~1.01, P = 0.97). Multivariate MR further identified that the partial effect of SUA on DN may be mediated by physical activity, low density lipoprotein cholesterol (LDL-C), insulin resistance (IR), and alcohol use.

Conclusion

The study establishes a causal link between elevated SUA levels and an increased risk of DN, with no evidence for a reverse association. This underscores the need for a comprehensive strategy in DN management, integrating urate-lowering interventions with modulations of the aforementioned mediators.

Protective effect of uridine on atrial fibrillation: a Mendelian randomisation study

Uridine, a pyrimidine nucleoside, is crucial in the synthesis of metabolites. According to observational studies, a higher plasma uridine level is associated with a lower risk of atrial fibrillation (AF). However, the casual relationship between uridine and AF is still unknown. In this study, we used the Mendelian randomisation (MR) approach to explore causality. Three genetic variants associated with uridine were identified from the Metabolomics GWAS server (7824 participants); summary-level datasets associated with AF were acquired from a genome-wide association study (GWAS) meta-analysis with 1,030,836 European participants (60,620 AF cases). We duplicated the MR analyses using datasets from AF HRC studies and the FinnGen Consortium, and then conducted a meta-analysis which combined the main results. The risk of AF was significantly associated with the genetically determined plasma uridine level (odds ratio [OR] 0.27; 95% confidence interval [CI] 0.16, 0.47; p = 2.39 × 10-6). The association remained consistent in the meta-analysis of the various datasets (OR 0.27; 95% CI 0.17, 0.42; p = 1.34 × 10-8). In conclusion, the plasma uridine level is inversely associated with the risk of AF. Raising the plasma uridine level may have prophylactic potential against AF.

Remote sensing-enabled machine learning for river water quality modeling under multidimensional uncertainty

Two fundamental problems have inhibited progress in the simulation of river water quality under climate (and other) uncertainty: 1) insufficient data, and 2) the inability of existing models to account for the complexity of factors (e.g., hydro-climatic, basin characteristics, land use features) affecting river water quality. To address these concerns this study presents a technique for augmenting limited ground-based observations of water quality variables with remote-sensed surface reflectance data by leveraging a machine learning model capable of accommodating the multidimensionality of water quality influences. Total Suspended Solids (TSS) can serve as a surrogate for chemical and biological pollutants of concern in surface water bodies. Historically, TSS data collection in the United States has been limited to the location of water treatment plants where state or federal agencies conduct regularly-scheduled water sampling. Mathematical models relating riverine TSS concentration to the explanatory factors have therefore been limited and the relationships between climate extremes and water contamination events have not been effectively diagnosed. This paper presents a method to identify these issues by utilizing a Long Short-Term Memory Network (LSTM) model trained on Moderate Resolution Imaging Spectroradiometer (MODIS) satellite reflectance data, which is calibrated to TSS data collected by the Ohio River Valley Water Sanitation Commission (ORSANCO). The methodology developed enables a thorough empirical analysis and data-driven algorithms able to account for spatial variability within the watershed and provide effective water quality prediction under uncertainty.

Predicting heroin potency from the analysis of paraphernalia: A tool for overdose prevention projects

In recent years, fatal and non-fatal heroin-related overdoses have increased in northeastern Italy, and the change in potency of heroin available at street level has been identified as a prominent factor associated with acute toxicity. Two very different products, high-potency and low-potency heroin were becoming available on the street, and no clear morphological characteristics could be used to easily distinguish them. A theoretical model for predicting heroin potency from rapid analysis of cigarette filters was developed as part of an overdose prevention project. The model was derived from the analysis of real heroin samples and exploits the common presence of caffeine in heroin as an adulterant. It was tested on laboratory prepared filters, real filters used to prepare heroin injections, and other paraphernalia. The model showed strong predictive ability and was used to implement a rapid alert system to inform drug users and healthcare institutions about the potency of heroin or other psychoactive substances circulating in the area. Cigarette filters were used as standard material, but other paraphernalia were successfully tested. The developed model is a dynamic tool whose parameters can be updated according to the market characteristics, so it can be useful for laboratories involved in drug analysis and similar prevention programs.

Screening Plasma Proteins for the Putative Drug Targets for Carpal Tunnel Syndrome

Carpal tunnel syndrome (CTS) is one of the most common work-related musculoskeletal disorders. The present study sought to identify putative causal proteins for CTS. We conducted a two-sample Mendelian randomization (MR) analysis to evaluate the causal association between 2859 plasma proteins (N = 35,559) and CTS (N = 1,239,680) based on the published GWAS summary statistics. Then we replicated the significant associations using an independent plasma proteome GWAS (N = 10,708). Sensitivity analyses were conducted to validate the robustness of MR results. Multivariate MR and mediation analyses were conducted to evaluate the mediation effects of body mass index (BMI), type 2 diabetes (T2D), and arm tissue composition on the association between putative causal proteins and CTS. Colocalization analysis was used to examine whether the identified proteins and CTS shared causal variant(s). Finally, we evaluated druggability of the identified proteins. Ten plasma proteins were identified as putative causal markers for CTS, including sCD14, PVR, LTOR3, CTSS, SIGIRR, IFNL3, ASPN, TM11D, ASIP, and ITIH1. Sensitivity analyses and reverse MR analysis validated the robustness of their causal effects. Arm tissue composition, BMI, and T2D may play a fully/partial mediating role in the causal relationships of ASIP, TM11D, IFNL3, PVR, and LTOR3 with CTS. The association of ASPN and sCD14 with CTS were supported by colocalization analysis. Druggability assessment demonstrated that sCD14, CTSS, TM11D, and IFNL3 were potential drug therapeutic targets. The present study identified several potential plasma proteins that were causally associated with CTS risk, providing new insights into the pathogenesis of protein-mediated CTS and offering potential targets for new therapies.

Optimizing sleep staging on multimodal time series: Leveraging borderline synthetic minority oversampling technique and supervised convolutional contrastive learning

Sleep is an important research area in nutritional medicine that plays a crucial role in human physical and mental health restoration. It can influence diet, metabolism, and hormone regulation, which can affect overall health and well-being. As an essential tool in the sleep study, the sleep stage classification provides a parsing of sleep architecture and a comprehensive understanding of sleep patterns to identify sleep disorders and facilitate the formulation of targeted sleep interventions. However, the class imbalance issue is typically salient in sleep datasets, which severely affects classification performances. To address this issue and to extract optimal multimodal features of EEG, EOG, and EMG that can improve the accuracy of sleep stage classification, a Borderline Synthetic Minority Oversampling Technique (B-SMOTE)-Based Supervised Convolutional Contrastive Learning (BST-SCCL) is proposed, which can avoid the risk of data mismatch between various sleep knowledge domains (varying health conditions and annotation rules) and strengthening learning characteristics of the N1 stage from the pair-wise segments comparison strategy. The lightweight residual network architecture with a novel truncated cross-entropy loss function is designed to accommodate multimodal time series and boost the training speed and performance stability. The proposed model has been validated on four well-known public sleep datasets (Sleep-EDF-20, Sleep-EDF-78, ISRUC-1, and ISRUC-3) and its superior performance (overall accuracy of 91.31-92.34%, MF1 of 88.21-90.08%, and Cohen's Kappa coefficient k of 0.87-0.89) has further demonstrated its effectiveness. It shows the great potential of contrastive learning for cross-domain knowledge interaction in precision medicine.

Machine learning strategy for identifying altered gut microbiomes for diagnostic screening in myasthenia gravis

Myasthenia gravis (MG) is a neuromuscular junction disease with a complex pathophysiology and clinical variation for which no clear biomarker has been discovered. We hypothesized that because changes in gut microbiome composition often occur in autoimmune diseases, the gut microbiome structures of patients with MG would differ from those without, and supervised machine learning (ML) analysis strategy could be trained using data from gut microbiota for diagnostic screening of MG. Genomic DNA from the stool samples of MG and those without were collected and established a sequencing library by constructing amplicon sequence variants (ASVs) and completing taxonomic classification of each representative DNA sequence. Four ML methods, namely least absolute shrinkage and selection operator, extreme gradient boosting (XGBoost), random forest, and classification and regression trees with nested leave-one-out cross-validation were trained using ASV taxon-based data and full ASV-based data to identify key ASVs in each data set. The results revealed XGBoost to have the best predicted performance. Overlapping key features extracted when XGBoost was trained using the full ASV-based and ASV taxon-based data were identified, and 31 high-importance ASVs (HIASVs) were obtained, assigned importance scores, and ranked. The most significant difference observed was in the abundance of bacteria in the Lachnospiraceae and Ruminococcaceae families. The 31 HIASVs were used to train the XGBoost algorithm to differentiate individuals with and without MG. The model had high diagnostic classification power and could accurately predict and identify patients with MG. In addition, the abundance of Lachnospiraceae was associated with limb weakness severity. In this study, we discovered that the composition of gut microbiomes differed between MG and non-MG subjects. In addition, the proposed XGBoost model trained using 31 HIASVs had the most favorable performance with respect to analyzing gut microbiomes. These HIASVs selected by the ML model may serve as biomarkers for clinical use and mechanistic study in the future. Our proposed ML model can identify several taxonomic markers and effectively discriminate patients with MG from those without with a high accuracy, the ML strategy can be applied as a benchmark to conduct noninvasive screening of MG.

Association between cumulative exposure periods of flupentixol or any antipsychotics and risk of lung cancer

BACKGROUND

Preclinical evidence suggests that certain antipsychotic medications may inhibit the development of lung cancer. This study aims to investigate the association between incident lung cancer and different cumulative exposure periods of flupentixol or any antipsychotics.

METHODS

Using electronic health records from the Hospital Authority in Hong Kong, this nested case-control study included case participants aged 18 years or older with newly diagnosed lung cancer after initiating antipsychotics between January 1, 2003, and August 31, 2022. Each case was matched to up to ten controls of the same sex and age, who were also antipsychotic users. Multivariable conditional logistic regression models were conducted to quantify the association between lung cancer and different cumulative exposure times of flupentixol (0-365 days [ref]; 366-1825 days; 1826+ days) and any antipsychotics (1-365 days [ref]; 366-1825 days; 1826+ days), separately.

RESULTS

Here we show that among 6435 cases and 64,348 matched controls, 64.06% are males, and 52.98% are aged 65-84 years. Compared to patients with less than 365 days of exposure, those with 366-1825 days of exposure to flupentixol (OR = 0.65 [95% CI, 0.47-0.91]) and any antipsychotics (0.42 [0.38-0.45]) have a lower risk of lung cancer. A decreased risk is observed in patients who have 1826+ days of cumulative use of any antipsychotics (0.54 [0.47-0.60]).

CONCLUSIONS

A reduced risk of lung cancer is observed in patients with more than one year of exposure to flupentixol or any antipsychotics. Further research on the association between lung cancer and other antipsychotic agents is warranted.

Digital image analysis and machine learning-assisted prediction of neoadjuvant chemotherapy response in triple-negative breast cancer

Background

Pathological complete response (pCR) is associated with favorable prognosis in patients with triple-negative breast cancer (TNBC). However, only 30-40% of TNBC patients treated with neoadjuvant chemotherapy (NAC) show pCR, while the remaining 60-70% show residual disease (RD). The role of the tumor microenvironment (TME) in NAC response in patients with TNBC remains unclear. In this study, we developed a machine learning-based two-step pipeline to distinguish between various histological components in hematoxylin and eosin (H&E)-stained whole slide images (WSIs) of TNBC tissue biopsies and to identify histological features that can predict NAC response.

Methods

H&E-stained WSIs of treatment-naïve biopsies from 85 patients (51 with pCR and 34 with RD) were separated through a stratified 8-fold cross validation strategy for the first step and leave one out cross validation strategy for the second step. A tile-level histology label prediction pipeline and four machine learning classifiers were used to analyze 468,043 tiles of WSIs. The best-trained classifier used 55 texture features from each tile to produce a probability profile during testing. The predicted histology classes were used to generate a histology classification map of the spatial distributions of different tissue regions. A patient-level NAC response prediction pipeline was trained with features derived from paired histology classification maps. The top graph-based features capturing the relevant spatial information across the different histological classes were provided to the radial basis function kernel support vector machine (rbfSVM) classifier for NAC treatment response prediction.

Results

The tile-level prediction pipeline achieved 86.72% accuracy for histology class classification, while the patient-level pipeline achieved 83.53% NAC response (pCR vs. RD) prediction accuracy. The histological class pairs with the strongest NAC response predictive ability were tumor and tumor tumor-infiltrating lymphocytes for pCR and microvessel density and polyploid giant cancer cells for RD.

Conclusion

Our machine learning pipeline can robustly identify clinically relevant histological classes that predict NAC response in TNBC patients and may help guide patient selection for NAC treatment.

Causal association between serum 25-Hydroxyvitamin D levels and cutaneous melanoma: a two-sample Mendelian randomization study

Background

Despite numerous observational studies on the association between serum 25-Hydroxyvitamin D levels and cutaneous melanoma, causal inferences remain ambiguous due to confounding and reverse causality. This study aimed to elucidate the causal relationship between serum 25-Hydroxyvitamin D levels and melanoma incidence using Mendelian randomization (MR).

Methods

A two-sample MR was conducted using genetic variants associated with serum 25-Hydroxyvitamin D levels as instrumental variables. Summary statistics for these variants were derived from genome-wide association studies, and those for melanoma risk were obtained from a comprehensive melanoma case-control study. Robustness of the results was assessed through sensitivity analyses, including the "leave-one-out" approach and tests for potential pleiotropy.

Results

The MR analysis provided substantial evidence of a positive causal relationship between serum 25-Hydroxyvitamin D levels and the incidence of cutaneous melanoma, suggesting that each unit increase in serum 25-Hydroxyvitamin D levels corresponds with an increased risk of melanoma. Tests for pleiotropy showed minimal effects, and the sensitivity analysis confirmed no disproportionate influence by any individual single nucleotide polymorphism (SNP).

Conclusion

The findings indicated a potentially causal positive association between serum 25-Hydroxyvitamin D levels and melanoma risk, challenging traditional beliefs about vitamin D's role in melanoma. This emphasizes the need for a balanced and personalized approach to vitamin D supplementation and sun exposure, particularly in high-risk populations. These results should be interpreted with caution due to potential unrecognized pleiotropy and confounding factors. Future research should focus on validating these findings in diverse populations and exploring underlying biological mechanisms.

Dimensions of elongate mineral particles and cancer: A review

CONTEXT

Based on a decade-long exploration, dimensions of elongate mineral particles are implicated as a pivotal component of their carcinogenic potency. This paper summarizes current understanding of the discovered relationships and their importance to the protection of public health.

OBJECTIVES

To demonstrate the relationships between cancer risk and dimensions (length, width, and other derivative characteristics) of mineral fibers by comparing the results and conclusions of previously published studies with newly published information.

METHODS

A database including 59 datasets comprising 341,949 records were utilized to characterize dimensions of elongate particles. The descriptive statistics, correlation and regression analysis, combined with Monte Carlo simulation, were used to select dimensional characteristics most relevant for mesothelioma and lung cancer risk prediction.

RESULTS

The highest correlation between mesothelioma potency factor and weight fraction of size categories is achieved for fibers with lengths >5.6 μm and widths ≤0.26 μm (R = 0.94, P < 0.02); no statistically significant potency was found for lengths <5 μm. These results are consistent with early published estimations, though are derived from a different approach. For combinations of amphiboles and chrysotile (with a consideration of a correction factor between mineral classes), the potency factors correlated most highly with a fraction of fibers longer than 5 μm and thinner than 0.2 μm for mesothelioma, and longer than 5 μm and thinner than 0.3 μm for lung cancer. Because the proportion of long, thin particles in asbestiform vs. non-asbestiform dusts is higher, the cancer potencies of the former are predicted at a significantly higher level. The analysis of particle dimensionality in human lung burden demonstrates positive selection for thinner fibers (especially for amosite and crocidolite) and prevailing fraction of asbestiform habit.

CONCLUSION

Dimensions of mineral fibers can be confirmed as one of the main drivers of their carcinogenicity. The width of fibers emerges as a primary potency predictor, and fibers of all widths with lengths shorter than 5 μm seem to be non-impactful for cancer risk. The mineral dust with a fibrous component is primarily carcinogenic if it contains amphibole fibers longer than 5 μm and thinner than 0.25 μm.

Hippocampal and Medial Prefrontal Cortex Fractal Spiking Patterns Encode Episodes and Rules

A central question in neuroscience is how the brain represents and processes information to guide behavior. The principles that organize brain computations are not fully known, and could include scale-free, or fractal patterns of neuronal activity. Scale-free brain activity may be a natural consequence of the relatively small subsets of neuronal populations that respond to task features, i.e., sparse coding. The size of the active subsets constrains the possible sequences of inter-spike intervals (ISI), and selecting from this limited set may produce firing patterns across wide-ranging timescales that form fractal spiking patterns. To investigate the extent to which fractal spiking patterns corresponded with task features, we analyzed ISIs in simultaneously recorded populations of CA1 and medial prefrontal cortical (mPFC) neurons in rats performing a spatial memory task that required both structures. CA1 and mPFC ISI sequences formed fractal patterns that predicted memory performance. CA1 pattern duration, but not length or content, varied with learning speed and memory performance whereas mPFC patterns did not. The most common CA1 and mPFC patterns corresponded with each region's cognitive function: CA1 patterns encoded behavioral episodes which linked the start, choice, and goal of paths through the maze whereas mPFC patterns encoded behavioral "rules" which guided goal selection. mPFC patterns predicted changing CA1 spike patterns only as animals learned new rules. Together, the results suggest that CA1 and mPFC population activity may predict choice outcomes by using fractal ISI patterns to compute task features.

A structural magnetic resonance imaging review of clinical motor outcomes from deep brain stimulation in movement disorders

Patients with movement disorders treated by deep brain stimulation do not always achieve successful therapeutic alleviation of motor symptoms, even in cases where surgery is without complications. Magnetic resonance imaging (MRI) offers methods to investigate structural brain-related factors that may be predictive of clinical motor outcomes. This review aimed to identify features which have been associated with variability in clinical post-operative motor outcomes in patients with Parkinson's disease, dystonia, and essential tremor from structural MRI modalities. We performed a literature search for articles published between 1 January 2000 and 1 April 2022 and identified 5197 articles. Following screening through our inclusion criteria, we identified 60 total studies (39 = Parkinson's disease, 11 = dystonia syndromes and 10 = essential tremor). The review captured a range of structural MRI methods and analysis techniques used to identify factors related to clinical post-operative motor outcomes from deep brain stimulation. Morphometric markers, including volume and cortical thickness were commonly identified in studies focused on patients with Parkinson's disease and dystonia syndromes. Reduced metrics in basal ganglia, sensorimotor and frontal regions showed frequent associations with reduced motor outcomes. Increased structural connectivity to subcortical nuclei, sensorimotor and frontal regions was also associated with greater motor outcomes. In patients with tremor, increased structural connectivity to the cerebellum and cortical motor regions showed high prevalence across studies for greater clinical motor outcomes. In addition, we highlight conceptual issues for studies assessing clinical response with structural MRI and discuss future approaches towards optimizing individualized therapeutic benefits. Although quantitative MRI markers are in their infancy for clinical purposes in movement disorder treatments, structural features obtained from MRI offer the powerful potential to identify candidates who are more likely to benefit from deep brain stimulation and provide insight into the complexity of disorder pathophysiology.

Two-sample Mendelian randomization analysis evaluates causal associations between inflammatory bowel disease and osteoporosis

Introduction

Over the past few years, multiple observational studies have speculated a potential association between inflammatory bowel disease (IBD), which includes ulcerative colitis (UC) and Crohn's disease (CD), and osteoporosis. However, no consensus has been reached regarding their interdependence and pathogenesis. Herein, we sought to further explore the causal associations between them.

Methods

We validated the association between IBD and reduced bone mineral density in humans based on genome-wide association studies (GWAS) data. To investigate the causal relationship between IBD and osteoporosis, we performed a two-sample Mendelian randomization study using training and validation sets. Genetic variation data for IBD, CD, UC, and osteoporosis were derived from published genome-wide association studies in individuals of European ancestry. After a series of robust quality control steps, we included eligible instrumental variables (SNPs) significantly associated with exposure (IBD/CD/UC). We adopted five algorithms, including MR Egger, Weighted median, Inverse variance weighted, Simple mode, and Weighted mode, to infer the causal association between IBD and osteoporosis. In addition, we evaluated the robustness of Mendelian randomization analysis by heterogeneity test, pleiotropy test, leave-one-out sensitivity test, and multivariate Mendelian randomization.

Results

Genetically predicted CD was positively associated with osteoporosis risk, with ORs of 1.060 (95% CIs 1.016, 1.106; p = 0.007) and 1.044 (95% CIs 1.002, 1.088; p = 0.039) for CD in the training and validation sets, respectively. However, Mendelian randomization analysis did not reveal a significant causal relationship between UC and osteoporosis (p > 0.05). Furthermore, we found that overall IBD was associated with osteoporosis prediction, with ORs of 1.050 (95% CIs 0.999, 1.103; p = 0.055) and 1.063 (95% CIs 1.019, 1.109; p = 0.005) in the training and validation sets, respectively.

Conclusion

We demonstrated the causal association between CD and osteoporosis, complementing the framework for genetic variants that predispose to autoimmune disease.

Goal Choices Modify Frontotemporal Memory Representations

Adapting flexibly to changing circumstances is guided by memory of past choices, their outcomes in similar circumstances, and a method for choosing among potential actions. The hippocampus (HPC) is needed to remember episodes, and the prefrontal cortex (PFC) helps guide memory retrieval. Single-unit activity in the HPC and PFC correlates with such cognitive functions. Previous work recorded CA1 and mPFC activity as male rats performed a spatial reversal task in a plus maze that requires both structures, found that PFC activity helps reactivate HPC representations of pending goal choices but did not describe frontotemporal interactions after choices. We describe these interactions after choices here. CA1 activity tracked both current goal location and the past starting location of single trials; PFC activity tracked current goal location better than past start location. CA1 and PFC reciprocally modulated representations of each other both before and after goal choices. After choices, CA1 activity predicted changes in PFC activity in subsequent trials, and the magnitude of this prediction correlated with faster learning. In contrast, PFC start arm activity more strongly modulated CA1 activity after choices correlated with slower learning. Together, the results suggest post-choice HPC activity conveys retrospective signals to the PFC, which combines different paths to common goals into rules. In subsequent trials, prechoice mPFC activity modulates prospective CA1 signals informing goal selection.SIGNIFICANCE STATEMENT HPC and PFC activity supports cognitive flexibility in changing circumstances. HPC signals represent behavioral episodes that link the start, choice, and goal of paths. PFC signals represent rules that guide goal-directed actions. Although prior studies described HPC-PFC interactions preceding decisions in the plus maze, post-decision interactions were not investigated. Here, we show post-choice HPC and PFC activity distinguished the start and goal of paths, and CA1 signaled the past start of each trial more accurately than mPFC. Postchoice CA1 activity modulated subsequent PFC activity, so rewarded actions were more likely to occur. Together, the results show that in changing circumstances, HPC retrospective codes modulate subsequent PFC coding, which in turn modulates HPC prospective codes that predict choices.

Validating Healthy Eating Index, Glycemic Index, and Glycemic Load with Modern Diets for E-Health Era

Predictors of healthy eating parameters, including the Healthy Eating Index (HEI), Glycemic Index (GI), and Glycemic Load (GL), were examined using various modern diets (n = 131) in preparation for personalized nutrition in the e-health era. Using Nutrition Data Systems for Research computerized software and artificial intelligence machine-learning-based predictive validation analyses, we included domains of HEI, caloric source, and various diets as the potentially modifiable factors. HEI predictors included whole fruits and whole grains, and empty calories. Carbohydrates were the common predictor for both GI and GL, with total fruits and Mexican diets being additional predictors for GI. The median amount of carbohydrates to reach an acceptable GL < 20 was predicted as 33.95 g per meal (median: 3.59 meals daily) with a regression coefficient of 37.33 across all daily diets. Diets with greater carbohydrates and more meals needed to reach acceptable GL < 20 included smoothies, convenient diets, and liquids. Mexican diets were the common predictor for GI and carbohydrates per meal to reach acceptable GL < 20; with smoothies (12.04), high-school (5.75), fast-food (4.48), Korean (4.30), Chinese (3.93), and liquid diets (3.71) presenting a higher median number of meals. These findings could be used to manage diets for various populations in the precision-based e-health era.

Developing a Prediction Model of Demolition-Waste Generation-Rate via Principal Component Analysis

Construction and demolition waste accounts for a sizable proportion of global waste and is harmful to the environment. Its management is therefore a key challenge in the construction industry. Many researchers have utilized waste generation data for waste management, and more accurate and efficient waste management plans have recently been prepared using artificial intelligence models. Here, we developed a hybrid model to forecast the demolition-waste-generation rate in redevelopment areas in South Korea by combining principal component analysis (PCA) with decision tree, k-nearest neighbors, and linear regression algorithms. Without PCA, the decision tree model exhibited the highest predictive performance (R2 = 0.872) and the k-nearest neighbors (Chebyshev distance) model exhibited the lowest (R2 = 0.627). The hybrid PCA-k-nearest neighbors (Euclidean uniform) model exhibited significantly better predictive performance (R2 = 0.897) than the non-hybrid k-nearest neighbors (Euclidean uniform) model (R2 = 0.664) and the decision tree model. The mean of the observed values, k-nearest neighbors (Euclidean uniform) and PCA-k-nearest neighbors (Euclidean uniform) models were 987.06 (kg·m-2), 993.54 (kg·m-2) and 991.80 (kg·m-2), respectively. Based on these findings, we propose the k-nearest neighbors (Euclidean uniform) model using PCA as a machine-learning model for demolition-waste-generation rate predictions.

Causal association between vitamin D and diabetic neuropathy: a Mendelian randomization analysis

OBJECTIVES

Vitamin D has been linked to diabetic neuropathy (DN) in previous epidemiological observational studies, however, their findings are inconsistent. The causal relationship between vitamin D and DN remains unknown. In this study we aim to investigate the causal association of serum 25-hydroxyvitamin D (25OHD) and DN.

METHODS

Based on summary statistics from publicly available genome-wide association studies (GWAS) database, we detected the genetic correlation between serum 25OHD levels and DN by a two-sample Mendelian randomization (MR) analysis. The inverse-variance weighted (IVW) method was used as the primary analysis, weighted median and MR-Egger were applied as complementary methods for MR estimates. In addition, we took sensitivity analyses including Cochran's Q test, MR-Pleiotropy Residual Sum and Outlier (MR-PRESSO) and leave-one-out analysis to ensure that we obtained stable and reliable results.

RESULTS

Our MR study showed no significant genetic association between serum 25OHD levels and DN (OR = 1.13, 95% CI = 0.81-1.57, P = 0.46). Furthermore, in the reverse direction analysis, we did not find a significant causal effect of DN and serum 25OHD levels (OR = 0.99, 95% CI = 0.98-1.00, P = 0.09). Results of MR-Egger, Weighted Median were consistent with those of the IVW method. The sensitivity analysis suggesting that no significant heterogeneity and genetic pleiotropy was observed.

CONCLUSIONS

Our results provided no evidence to support the causal association of serum 25OHD levels with DN.

Development of Machine Learning Model for Prediction of Demolition Waste Generation Rate of Buildings in Redevelopment Areas

Owing to a rapid increase in waste, waste management has become essential, for which waste generation (WG) information has been effectively utilized. Various studies have recently focused on the development of reliable predictive models by applying artificial intelligence to the construction and prediction of WG information. In this study, research was conducted on the development of machine learning (ML) models for predicting the demolition waste generation rate (DWGR) of buildings in redevelopment areas in South Korea. Various ML algorithms (i.e., artificial neural network (ANN), K-nearest neighbors (KNN), linear regression (LR), random forest (RF), and support vector machine (SVM)) were applied to the development of an optimal predictive model, and the main hyper parameters (HPs) for each algorithm were optimized. The results suggest that ANN-ReLu (coefficient of determination (R²) 0.900, the ratio of percent deviation (RPD) 3.16), SVM-polynomial (R² 0.889, RPD 3.00), and ANN-logistic (R² 0.883, RPD 2.92) are the best ML models for predicting the DWGR. They showed average errors of 7.3%, 7.4%, and 7.5%, respectively, compared to the average observed values, confirming the accurate predictive performance, and in the uncertainty analysis, the d-factor of the models appeared less than 1, showing that the presented models are reliable. Through a comparison with ML algorithms and HPs applied in previous related studies, the results herein also showed that the selection of various ML algorithms and HPs is important in developing optimal ML models for WG management.

Genetically supported causality between benign prostate hyperplasia and urinary bladder neoplasms: A mendelian randomization study

Background: Observational studies have suggested a possible association between benign prostate hyperplasia (BPH) and bladder cancer (BLCA). However, these studies are prone to errors and limitations or confounding factors, making them unsuitable for assessing the causal relationship between BPH and BLCA.

Objective: Two-sample Mendelian randomization (MR) was performed to determine a possible association between genetically predicted BPH and the risk of BLCA.

Methods: A two-sample MR analysis was performed utilizing the Integrative Epidemiology Unit genome-wide association (GWAS) database of the Medical Research Council, United Kingdom A series of control steps, including five primary methods, were performed to identify the most suitable instrumental variables (IVs) for MR analysis. Sensitivity analysis was conducted to avoid statistical errors, including heterogeneity and pleiotropic bias.

Results: Genetic variants associated with BPH (P &lt; 5 × 10–8) and BLCA (P &lt; 5 × 10–6) were identified as instrumental variables and assessed using GWAS summary data (BPH, 4,670 cases vs. 458,340 controls; BLCA, 1,279 cases vs. 372,016 controls). BPH exhibited a positive effect on the occurrence of BLCA (inverse variance weighted (IVW), odds ratio (OR) = 1.095, 95% confidence interval (CI) = 1.030–1.165, p = 0.003), but there was no causal effect for BLCA on BPH (IVW, OR = 1.092, 95% CI = 0.814–1.465, p = 0.554).

Conclusion: Genetically predicted BPH was associated with a higher risk of BLCA in all histological subtypes. In contrast, the evidence was not significant to back the causality of genetically induced BLCA on BPH. These findings indicate that BPH plays a key role in developing BLCA in the European population. Further studies are needed to uncover the underlying mechanisms.

Drug repurposing using meta-analysis of gene expression in Alzheimer's disease

INTRODUCTION

Alzheimer's disease and other forms of dementia are disease that bring an increased global burden. However, the medicine developed to date remains limited. The purpose of this study is to predict drug repositioning candidates using a computational method that integrates gene expression profiles on Alzheimer's disease and compound-induced changes in gene expression levels.

METHODS

Gene expression data on Alzheimer's disease were obtained from the Gene Expression Omnibus (GEO) and we conducted a meta-analysis of their gene expression levels. The reverse scores of compound-induced gene expressions were computed based on the reversal relationship between disease and drug gene expression profiles.

RESULTS

Reversal genes and the candidate compounds were identified by the leave-one-out cross-validation procedure. Additionally, the half-maximal inhibitory concentration (IC50) values and the blood-brain barrier (BBB) permeability of candidate compounds were obtained from ChEMBL and PubChem, respectively.

CONCLUSION

New therapeutic target genes and drug candidates against Alzheimer's disease were identified by means of drug repositioning.

Oxidative stress markers-driven prognostic model to predict post-discharge mortality in heart failure with reduced ejection fraction

Background

Current predictive models based on biomarkers reflective of different pathways of heart failure with reduced ejection fraction (HFrEF) pathogenesis constitute a useful tool for predicting death risk among HFrEF patients. The purpose of the study was to develop a new predictive model for post-discharge mortality risk among HFrEF patients, based on a combination of clinical patients’ characteristics, N-terminal pro-B-type Natriuretic peptide (NT-proBNP) and oxidative stress markers as a potentially valuable tool for routine clinical practice.

Methods

116 patients with stable HFrEF were recruited in a prospective single-center study. Plasma levels of NT-proBNP and oxidative stress markers [superoxide dismutase (SOD), glutathione peroxidase (GPX), uric acid (UA), total bilirubin (TB), gamma-glutamyl transferase (GGT) and total antioxidant capacity (TAC)] were measured in the stable predischarge condition. Generalized linear model (GLM), random forest and extreme gradient boosting models were developed to predict post-discharge mortality risk using clinical and laboratory data. Through comprehensive evaluation, the most performant model was selected.

Results

During a median follow-up of 525 days (7–930), 33 (28%) patients died. Among the three created models, the GLM presented the best performance for post-discharge death prediction in HFrEF. The predictors included in the GLM model were age, female sex, beta blockers, NT-proBNP, left ventricular ejection fraction (LVEF), TAC levels, admission systolic blood pressure (SBP), angiotensin-converting enzyme inhibitors/angiotensin receptor II blockers (ACEI/ARBs) and UA levels. Our model had a good discriminatory power for post-discharge mortality [The area under the curve (AUC) = 74.5%]. Based on the retained model, an online calculator was developed to allow the identification of patients with heightened post-discharge death risk.

Conclusion

In conclusion, we created a new and simple tool that may allow the identification of patients at heightened post-discharge mortality risk and could assist the treatment decision-making.

Increased interferon-γ levels and risk of severe malaria: a meta-analysis

Interferon (IFN)-γ contributes to the pathogenesis of severe malaria; however, its mechanism remains unclear. Herein, differences in IFN-γ levels between patients with severe and uncomplicated malaria were evaluated using qualitative and quantitative (meta-analysis) approaches. The systematic review protocol was registered at PROSPERO (ID: CRD42022315213). The searches for relevant studies were performed in five databases, including PubMed, Scopus, Embase, MEDLINE and Web of Science, between 1 January and 10 July 2022. A meta-analysis was conducted to pool the mean difference (MD) of IFN-γ levels between patients with severe malaria and those with uncomplicated malaria using a random-effects model (DerSimonian and Laird method). Overall, qualitative synthesis indicated that most studies (14, 58.3%) reported no statistically significant difference in IFN-γ levels between patients with severe malaria and those with uncomplicated malaria. Meanwhile, remaining studies (9, 37.5%) reported that IFN-γ levels were significantly higher in patients with severe malaria than those in patients with uncomplicated malaria. Only one study (4.17%) reported that IFN-γ levels were significantly lower in patients with severe malaria than those in patients with uncomplicated malaria. The meta-analysis results indicated that patients with severe malaria had higher mean IFN-γ levels than those with uncomplicated malaria (p < 0.001, MD: 13.63 pg/mL, 95% confidence interval: 6.98-20.29 pg/mL, I²: 99.02%, 14 studies/15 study sites, 652 severe cases/1096 uncomplicated cases). In summary, patients with severe malaria exhibited higher IFN-γ levels than those with uncomplicated malaria, although the heterogeneity of the outcomes is yet to be elucidated. To confirm whether alteration in IFN-γ levels of patients with malaria may indicate disease severity and/or poor prognosis, further studies are warranted.

Identification of biomarkers for hepatocellular carcinoma based on single cell sequencing and machine learning algorithms

Hepatocellular carcinoma (HCC) remains one of the most lethal cancers around the world. Precision oncology will be crucial for further improving the prognosis of HCC patients. Compared with traditional bulk RNA-seq, single-cell RNA sequencing (scRNA-seq) enables the transcriptomes of a great deal of individual cells assayed in an unbiased manner, showing the potential to deeply reveal tumor heterogeneity. In this study, based on the scRNA-seq results of primary neoplastic cells and paired normal liver cells from eight HCC patients, a new strategy of machine learning algorithms was applied to screen core biomarkers that distinguished HCC tumor tissues from the adjacent normal liver. Expression profiles of HCC cells and normal liver cells were first analyzed by maximum relevance minimum redundancy (mRMR) to get a top 50 signature gene feature. For further analysis, the incremental feature selection (IFS) method and leave-one-out cross validation (LOOCV) were conducted to build an optimal classification model and to extract 21 potentially essential biomarkers for HCC cells. Our results provided new insights into HCC pathogenesis that might be valuable for HCC diagnosis and therapy.

Advances in integrated genomic selection for rapid genetic gain in crop improvement: a review

Genomic selection and its importance in crop breeding. Integration of GS with new breeding tools and developing SOP for GS to achieve maximum genetic gain with low cost and time. The success of conventional breeding approaches is not sufficient to meet the demand of a growing population for nutritious food and other plant-based products. Whereas, marker assisted selection (MAS) is not efficient in capturing all the favorable alleles responsible for economic traits in the process of crop improvement. Genomic selection (GS) developed in livestock breeding and then adapted to plant breeding promised to overcome the drawbacks of MAS and significantly improve complicated traits controlled by gene/QTL with small effects. Large-scale deployment of GS in important crops, as well as simulation studies in a variety of contexts, addressed G × E interaction effects and non-additive effects, as well as lowering breeding costs and time. The current study provides a complete overview of genomic selection, its process, and importance in modern plant breeding, along with insights into its application. GS has been implemented in the improvement of complex traits including tolerance to biotic and abiotic stresses. Furthermore, this review hypothesises that using GS in conjunction with other crop improvement platforms accelerates the breeding process to increase genetic gain. The objective of this review is to highlight the development of an appropriate GS model, the global open source network for GS, and trans-disciplinary approaches for effective accelerated crop improvement. The current study focused on the application of data science, including machine learning and deep learning tools, to enhance the accuracy of prediction models. Present study emphasizes on developing plant breeding strategies centered on GS combined with routine conventional breeding principles by developing GS-SOP to achieve enhanced genetic gain.

Validating Accuracy of an Internet-Based Application against USDA Computerized Nutrition Data System for Research on Essential Nutrients among Social-Ethnic Diets for the E-Health Era

Internet-based applications (apps) are rapidly developing in the e-Health era to assess the dietary intake of essential macro-and micro-nutrients for precision nutrition. We, therefore, validated the accuracy of an internet-based app against the Nutrition Data System for Research (NDSR), assessing these essential nutrients among various social-ethnic diet types. The agreement between the two measures using intraclass correlation coefficients was good (0.85) for total calories, but moderate for caloric ranges outside of <1000 (0.75) and >2000 (0.57); and good (>0.75) for most macro- (average: 0.85) and micro-nutrients (average: 0.83) except cobalamin (0.73) and calcium (0.51). The app underestimated nutrients that are associated with protein and fat (protein: −5.82%, fat: −12.78%, vitamin B12: −13.59%, methionine: −8.76%, zinc: −12.49%), while overestimated nutrients that are associated with carbohydrate (fiber: 6.7%, B9: 9.06%). Using artificial intelligence analytics, we confirmed the factors that could contribute to the differences between the two measures for various essential nutrients, and they included caloric ranges; the differences between the two measures for carbohydrates, protein, and fat; and diet types. For total calories, as an example, the source factors that contributed to the differences between the two measures included caloric range (<1000 versus others), fat, and protein; for cobalamin: protein, American, and Japanese diets; and for folate: caloric range (<1000 versus others), carbohydrate, and Italian diet. In the e-Health era, the internet-based app has the capacity to enhance precision nutrition. By identifying and integrating the effects of potential contributing factors in the algorithm of output readings, the accuracy of new app measures could be improved.

Effect of meteorological conditions on leisure walking: a time series analysis and the application of outdoor thermal comfort indexes

Leisure walking is affected by meteorological conditions. However, it is still not clear what scales of meteorological conditions and thermal status affect the number of people who choose to leisure walk. Using a time series regression, this study examines the heat-leisure walking relationship by analyzing the effect of the seasons, weather, microclimate, and outdoor thermal comfort on walking count. Eight thermal indexes were selected to estimate the pedestrians' thermal comfort, and their predictive capacities in walking count were evaluated. Particular consideration was given to identifying heat thresholds of walking that determined the tolerance range of pedestrian heat stress. Four years of hourly daytime walking counts and publicly available ASOS meteorological data at Seoul-lo 7017, a pedestrian bridge in Seoul, were used for the analysis. Our findings indicate that walking count is correlated with seasonal climatic variations, with the highest number of pedestrians observed in fall and the lowest in summer. Moreover, air temperature played a vital role, showing that a 5.0 °C rise in temperature was associated with a 1.34% rise in the square root of the walking count. Its impact becomes greater when combined with intense solar radiation and higher absolute humidity. The heat threshold for walking was between 23.8 °C and 26.2 °C. Empirical model indexes showed the highest predictive capacity in walking count at approximately 30.0%, which was followed by rational model indexes at 28.0%.

Mendelian Randomization Analysis Reveals Causal Effects of Plasma Proteome on Body Composition Traits

CONTEXT

Observational studies have demonstrated associations between plasma proteins and obesity, but evidence of causal relationship remains to be studied.

OBJECTIVE

We aimed to evaluate the causal relationship between plasma proteins and body composition.

METHODS

We conducted a 2-sample Mendelian randomization (MR) analysis based on the genome-wide association study (GWAS) summary statistics of 23 body composition traits and 2656 plasma proteins. We then performed hierarchical cluster analysis to evaluate the structure and pattern of the identified causal associations, and we performed gene ontology enrichment analysis to explore the functional relevance of the identified proteins.

RESULTS

We identified 430 putatively causal effects of 96 plasma proteins on 22 body composition traits (except obesity status) with strong MR evidence (P < 2.53 × 10 - 6, at a Bonferroni-corrected threshold). The top 3 causal associations are follistatin (FST) on trunk fat-free mass (Beta = -0.63, SE = 0.04, P = 2.00 × 10-63), insulin-like growth factor-binding protein 1 (IGFBP1) on trunk fat-free mass (Beta = -0.54, SE = 0.03, P = 1.79 × 10-57) and r-spondin-3 (RSPO3) on WHR (waist circumference/hip circumference) (Beta = 0.01, SE = 4.47 × 10-4, P = 5.45 × 10-60), respectively. Further clustering analysis and pathway analysis demonstrated that the pattern of causal effect to fat mass and fat-free mass may be different.

CONCLUSION

Our findings may provide evidence for causal relationships from plasma proteins to various body composition traits and provide basis for further targeted functional studies.

Validating Accuracy of a Mobile Application against Food Frequency Questionnaire on Key Nutrients with Modern Diets for mHealth Era

In preparation for personalized nutrition, an accurate assessment of dietary intakes on key essential nutrients using smartphones can help promote health and reduce health risks across vulnerable populations. We, therefore, validated the accuracy of a mobile application (app) against Food Frequency Questionnaire (FFQ) using artificial intelligence (AI) machine-learning-based analytics, assessing key macro- and micro-nutrients across various modern diets. We first used Bland and Altman analysis to identify and visualize the differences between the two measures. We then applied AI-based analytics to enhance prediction accuracy, including generalized regression to identify factors that contributed to the differences between the two measures. The mobile app underestimated most macro- and micro-nutrients compared to FFQ (ranges: -5% for total calories, -19% for cobalamin, -33% for vitamin E). The average correlations between the two measures were 0.87 for macro-nutrients and 0.84 for micro-nutrients. Factors that contributed to the differences between the two measures using total calories as an example, included caloric range (1000-2000 versus others), carbohydrate, and protein; for cobalamin, included caloric range, protein, and Chinese diet. Future studies are needed to validate actual intakes and reporting of various diets, and to examine the accuracy of mobile App. Thus, a mobile app can be used to support personalized nutrition in the mHealth era, considering adjustments with sources that could contribute to the inaccurate estimates of nutrients.

Letter to the Editor: Epidemiology holds a key to the validation of toxicological models for elongate mineral particles

•

Human epidemiological data remains the main source for information about health effects of elongate mineral particles.

•

There are limitations in using animal data (in vitro and in vivo) for quantification of risk.

•

Toxicological modeling is one of the most promising tools to estimate risks for workers and communities.

The authors reply to the comments of Drs. Gualtieri and di Giuseppe on the short communication by Wylie and Korchevskiy – Carcinogenicity of fibrous glaucophane: how to fill data gaps? (2021 Current Research in Toxicology Volume 2, pp. 202–203). The role of epidemiology in establishing the toxicity of elongate mineral particles is emphasized. The validation of quantitative structure–activity relationship (QSAR) models by disease outcome is mentioned as one of the most important tools in advancing the new approaches in mineral particle toxicology.

Objective Assessment System for Hearing Prediction Based on Stimulus-Frequency Otoacoustic Emissions

Stimulus-frequency otoacoustic emissions (SFOAEs) can be useful tools for assessing cochlear function noninvasively. However, there is a lack of reports describing their utility in predicting hearing capabilities. Data for model training were collected from 245 and 839 ears with normal hearing and sensorineural hearing loss, respectively. Based on SFOAEs, this study developed an objective assessment system consisting of three mutually independent modules, with the routine test module and the fast test module used for threshold prediction and the hearing screening module for identifying hearing loss. Results evaluated via cross-validation show that the routine test module and the fast test module predict hearing thresholds with similar performance from 0.5 to 8 kHz, with mean absolute errors of 7.06–11.61 dB for the routine module and of 7.40–12.60 dB for the fast module. However, the fast module involves less test time than is needed in the routine module. The hearing screening module identifies hearing status with a large area under the receiver operating characteristic curve (0.912–0.985), high accuracy (88.4–95.9%), and low false negative rate (2.9–7.0%) at 0.5–8 kHz. The three modules are further validated on unknown data, and the results are similar to those obtained through cross-validation, indicating these modules can be well generalized to new data. Both the routine module and fast module are potential tools for predicting hearing thresholds. However, their prediction performance in ears with hearing loss requires further improvement to facilitate their clinical utility. The hearing screening module shows promise as a clinical tool for identifying hearing loss.

Dimensional determinants for the carcinogenic potency of elongate amphibole particles

CONTEXT

Carcinogenic properties of particulates depend, among other factors, on dimensional characteristics that affect their ability to reach sensitive tissue, to be removed or retained, and to interact with the cells.

OBJECTIVE

To model mesothelioma and lung cancer potency of amphibole particles based on their dimensional characteristics and mineral habit (asbestiform vs. nonasbestiform) utilizing epidemiological data and detailed size information.

METHODS

The datasets from recently created depository of dimensional information of elongate mineral particles were used to correlate mesothelioma and lung cancer potency with the fraction of particles in a specific size range and the ratio of length and width in different powers. In addition, the cancer potency factors were estimated and compared for 30 asbestiform, 15 nonasbestiform, and 10 mixed datasets.

RESULTS

For particles longer than 5 µm, the highest correlation with mesothelioma potency was achieved for width <0.22 µm, and with lung cancer <0.28 µm. The statistical power of the correlation was observed to lose significance at a maximum width of 0.6-0.7 µm. Mesothelioma potency correlated with length in the power of 1.9 divided by width in the power of 2.97, lung cancer potency with length in the power of 0.4 divided by width in the power of 1.17. The predicted cancer potencies of asbestiform, nonasbestiform, and mixed categories were significantly different.

CONCLUSION

While additional studies in this direction are warranted, this paper should serve as an additional confirmation for the role of fiber dimensions in the carcinogenicity of amphibole elongate mineral particles (EMPs).

Author response to Cunha et al

The need to identify biomarkers to predict immunotherapy response for rare cancers has been long overdue. We aimed to study this in our paper, 'Radiomics analysis for predicting pembrolizumab response in patients with advanced rare cancers'. In this response to the Letter to the Editor by Cunha et al, we explain and discuss the reasons behind choosing LASSO (Least Absolute Shrinkage and Selection Operator) and XGBoost (eXtreme Gradient Boosting) with LOOCV (Leave-One-Out Cross-Validation) as the feature selection and classifier method, respectively for our radiomics models. Also, we highlight what care was taken to avoid any overfitting on the models. Further, we checked for the multicollinearity of the features. Additionally, we performed 10-fold cross-validation instead of LOOCV to see the predictive performance of our radiomics models.