Avoiding common pitfalls when clustering biological data

Intra-clustering analysis reveals tissue-specific mutational patterns

BACKGROUND AND OBJECTIVE

The identification of tissue-specific mutational patterns associated with cancer is challenging due to the low frequency of certain mutations and the high variability among tumors within the same cancer type. To address the inter-tumoral heterogeneity issue, our study aims to uncover infrequent mutational patterns by proposing a novel intra-clustering analysis.

METHODS

A Network Graph of 8303 patients and 198 genes was constructed using single-point-mutation data from The Cancer Genome Atlas (TCGA). Patient-gene groups were retrieved with the parallel use of two separate methodologies based on the: (a) Barber's modularity index, and (b) network dynamics. An intra-clustering analysis was employed to explore the patterns within smaller patient subgroups in two phases: i) to determine the significant presence of a gene with a cancer type using the Fisher's exact test and ii) to determine gene-to-gene patterns using multiple correspondence analysis and DISCOVER. The results are followed by a Benjamini-Hochberg false discovery rate of 5%.

RESULTS

This analysis was applied over 24 statistically meaningful groups of 2619 patients spanning 21 cancer types and it recovered 42 mutational patterns that are not reported in the TCGA consortium publications. Notably, our findings: (i) suggest that AMER1 mutations are a putative separative element between colon and rectal adenocarcinomas, (ii) highlight the significant presence of RAC1 in head and neck squamous cell carcinoma (iii) suggest that EP300 mutations in head and neck squamous cell carcinoma are irrelevant of the HPV status of the patients and (iv) show that mutational-based clusters can contain patients with contrasting genetic alterations.

CONCLUSIONS

The proposed intra-clustering analysis extracted statistically significant relationships within clusters, uncovering putative clinically relevant connections and disentangling mutational heterogeneity.

A biological model of nonlinear dimensionality reduction

Obtaining appropriate low-dimensional representations from high-dimensional sensory inputs in an unsupervised manner is essential for straightforward downstream processing. Although nonlinear dimensionality reduction methods such as t-distributed stochastic neighbor embedding (t-SNE) have been developed, their implementation in simple biological circuits remains unclear. Here, we develop a biologically plausible dimensionality reduction algorithm compatible with t-SNE, which uses a simple three-layer feedforward network mimicking the Drosophila olfactory circuit. The proposed learning rule, described as three-factor Hebbian plasticity, is effective for datasets such as entangled rings and MNIST, comparable to t-SNE. We further show that the algorithm could be working in olfactory circuits in Drosophila by analyzing the multiple experimental data in previous studies. We lastly suggest that the algorithm is also beneficial for association learning between inputs and rewards, allowing the generalization of these associations to other inputs not yet associated with rewards.

Ecohydraulics-based environmental flow assessment in two arid North African rivers

North Africa is among the most water-stressed regions in the world; still, the habitat requirements of its freshwater biota are largely unknown. In this study, (i) we developed habitat suitability curves (HSCs) for freshwater macroinvertebrates in two poorly studied, regulated North African rivers (Ziz and Oum Er-Rbia), and (ii) assessed environmental flows downstream of each river dam by incorporating the HSCs in two-dimensional ecohydraulic models. We demonstrate a low-cost sampling methodology combined with freely distributed ecohydraulic modeling software. The results showed that macroinvertebrates in the arid-desert Ziz River could tolerate a wide range of habitats in terms of flow velocity and water depth compared to the arid-steppe Oum Er-Rbia River, probably due to their adaptation to extreme (arid-desert) environmental conditions. Optimal environmental flows downstream of the Al Hassan Addakhil (Ziz River) and the Al Massira (Oum Er-Rbia River) dams were 1 m3/s and 2 m3/s, respectively. However, environmental flows at 0.5 m3/s and 1 m3/s, respectively, could still maintain sustainable freshwater biota downstream of the dams. The results further highlight the critical status of the Ziz River, which was completely dry, and the alarming status of the Oum Er-Rbia River due to the significant reduction in the water levels of the Al Massira Dam. In a continuously changing climate, we suggest that the proposed environmental flows should be immediately delivered to prevent droughts and ensure healthy freshwater communities downstream of the dams, within a basin-wide freshwater management framework. In this water scarce region, more research is necessary to increase ecological awareness about these understudied freshwater systems and achieve a balance between human needs and ecosystem requirements.

Clustering affordable care act qualified health plans to understand how and where insurance facilitates or impedes access to HIV prevention

BACKGROUND

With access to and uptake of pre-exposure prophylaxis (PrEP), the United States can prevent new HIV infections. To end the HIV epidemic, health insurance plans must facilitate access to comprehensive preventive care benefits. Since plan benefit designs vary considerably by plan, it is difficult to systematically determine plans that facilitate and restrict preventive services for PrEP.

METHODS

We applied an unsupervised machine learning method to cluster 17,061 Qualified Health Plans offered to individuals. We examined the clusters to draw conclusions about the types of benefits insurance companies tend to group together in plans. Then we analyzed the geographic distribution of those clusters across the United States to assess geographic inequities in access to HIV preventive care.

RESULTS

Our method uncovered three cohesive clusters of plans. Plans in Cluster 1: the least restrictive cluster, facilitate access to preventive care using copays over coinsurance on almost all benefits; Cluster 2: the moderately restrictive cluster, plans cover HIV prevention benefits with copays but restrict access to general health benefits with coinsurance; and Cluster 3: the most restrictive cluster, plans cover almost all benefits using coinsurance. Overall, increased prior authorization requirements tend to accompany reductions in out-of-pocket costs. Examining the geographic plan distribution, states with at least one rating area where at least 75% of plans offered are in the most restrictive cluster included: Georgia, Illinois, Missouri, Oklahoma, Texas, Virginia, and Wyoming.

CONCLUSIONS

Insurance plan design is complex. To address the ambitious call to end the HIV epidemic in this country, plans should also take into account both public health and health equity factors to create plan designs that ensure access to critical preventive services for people who need them most. Addressing the growing disparities in PrEP access along racial and ethnic lines should be a national priority, and federal and state insurance regulators as well as insurance plans themselves should be part of the conversation about how to ensure people who would benefit from PrEP can access it. Better state/federal regulation of plan design to ensure access is consistent, equitable, and based on clinical recommendations will reduce the variability across plan designs.

AAclust: k-optimized clustering for selecting redundancy-reduced sets of amino acid scales

Summary

Amino acid scales are crucial for sequence-based protein prediction tasks, yet no gold standard scale set or simple scale selection methods exist. We developed AAclust, a wrapper for clustering models that require a pre-defined number of clusters k, such as k-means. AAclust obtains redundancy-reduced scale sets by clustering and selecting one representative scale per cluster, where k can either be optimized by AAclust or defined by the user. The utility of AAclust scale selections was assessed by applying machine learning models to 24 protein benchmark datasets. We found that top-performing scale sets were different for each benchmark dataset and significantly outperformed scale sets used in previous studies. Noteworthy is the strong dependence of the model performance on the scale set size. AAclust enables a systematic optimization of scale-based feature engineering in machine learning applications.

Availability and implementation

The AAclust algorithm is part of AAanalysis, a Python-based framework for interpretable sequence-based protein prediction, which is documented and accessible at https://aaanalysis.readthedocs.io/en/latest and https://github.com/breimanntools/aaanalysis.

AAontology: An Ontology of Amino Acid Scales for Interpretable Machine Learning

Amino acid scales are crucial for protein prediction tasks, many of them being curated in the AAindex database. Despite various clustering attempts to organize them and to better understand their relationships, these approaches lack the fine-grained classification necessary for satisfactory interpretability in many protein prediction problems. To address this issue, we developed AAontology-a two-level classification for 586 amino acid scales (mainly from AAindex) together with an in-depth analysis of their relations-using bag-of-word-based classification, clustering, and manual refinement over multiple iterations. AAontology organizes physicochemical scales into 8 categories and 67 subcategories, enhancing the interpretability of scale-based machine learning methods in protein bioinformatics. Thereby it enables researchers to gain a deeper biological insight. We anticipate that AAontology will be a building block to link amino acid properties with protein function and dysfunctions as well as aid informed decision-making in mutation analysis or protein drug design.

Long Non-Coding RNAs, Nuclear Receptors and Their Cross-Talks in Cancer-Implications and Perspectives

Long non-coding RNAs (lncRNAs) play key roles in various epigenetic and post-transcriptional events in the cell, thereby significantly influencing cellular processes including gene expression, development and diseases such as cancer. Nuclear receptors (NRs) are a family of ligand-regulated transcription factors that typically regulate transcription of genes involved in a broad spectrum of cellular processes, immune responses and in many diseases including cancer. Owing to their many overlapping roles as modulators of gene expression, the paths traversed by lncRNA and NR-mediated signaling often cross each other; these lncRNA-NR cross-talks are being increasingly recognized as important players in many cellular processes and diseases such as cancer. Here, we review the individual roles of lncRNAs and NRs, especially growth factor modulated receptors such as androgen receptors (ARs), in various types of cancers and how the cross-talks between lncRNAs and NRs are involved in cancer progression and metastasis. We discuss the challenges involved in characterizing lncRNA-NR associations and how to overcome them. Furthering our understanding of the mechanisms of lncRNA-NR associations is crucial to realizing their potential as prognostic features, diagnostic biomarkers and therapeutic targets in cancer biology.

eLIMS: Ensemble Learning-Based Spatial Segmentation of Mass Spectrometry Imaging to Explore Metabolic Heterogeneity

Spatial segmentation is an essential processing method for image analysis aiming to identify the characteristic suborgans or microregions from mass spectrometry imaging (MSI) data, which is critical for understanding the spatial heterogeneity of biological information and function and the underlying molecular signatures. Due to the intrinsic characteristics of MSI data including spectral nonlinearity, high-dimensionality, and large data size, the common segmentation methods lack the capability for capturing the accurate microregions associated with biological functions. Here we proposed an ensemble learning-based spatial segmentation strategy, named eLIMS, that combines a randomized unified manifold approximation and projection (r-UMAP) dimensionality reduction module for extracting significant features and an ensemble pixel clustering module for aggregating the clustering maps from r-UMAP. Three MSI datasets are used to evaluate the performance of eLIMS, including mouse fetus, human adenocarcinoma, and mouse brain. Experimental results demonstrate that the proposed method has potential in partitioning the heterogeneous tissues into several subregions associated with anatomical structure, i.e., the suborgans of the brain region in mouse fetus data are identified as dorsal pallium, midbrain, and brainstem. Furthermore, it effectively discovers critical microregions related to physiological and pathological variations offering new insight into metabolic heterogeneity.

Discovery and description of novel phage genomes from urban microbiomes sampled by the MetaSUB consortium

Bacteriophages are recognized as the most abundant members of microbiomes and have therefore a profound impact on microbial communities through the interactions with their bacterial hosts. The International Metagenomics and Metadesign of Subways and Urban Biomes Consortium (MetaSUB) has sampled mass-transit systems in 60 cities over 3 years using metagenomics, throwing light into these hitherto largely unexplored urban environments. MetaSUB focused primarily on the bacterial community. In this work, we explored MetaSUB metagenomic data in order to recover and analyze bacteriophage genomes. We recovered and analyzed 1714 phage genomes with size at least 40 kbp, from the class Caudoviricetes, the vast majority of which (80%) are novel. The recovered genomes were predicted to belong to temperate (69%) and lytic (31%) phages. Thirty-three of these genomes have more than 200 kbp, and one of them reaches 572 kbp, placing it among the largest phage genomes ever found. In general, the phages tended to be site-specific or nearly so, but 194 genomes could be identified in every city from which phage genomes were retrieved. We predicted hosts for 48% of the phages and observed general agreement between phage abundance and the respective bacterial host abundance, which include the most common nosocomial multidrug-resistant pathogens. A small fraction of the phage genomes are carriers of antibiotic resistance genes, and such genomes tended to be particularly abundant in the sites where they were found. We also detected CRISPR-Cas systems in five phage genomes. This study expands the previously reported MetaSUB results and is a contribution to the knowledge about phage diversity, global distribution, and phage genome content.

Cluster analysis as a clinical and research tool in Behçet's syndrome

PURPOSE OF REVIEW

The purpose of this review was to comprehensively summarize recent phenotype research findings in Behçet's syndrome.

RECENT FINDINGS

Cluster analysis has recently been employed as a phenotype research tool in Behçet's syndrome. Studies reported different clustering patterns caused by biological variation and some degree of artificial heterogeneity. However, some clusters were more consistent than others: 1) oral ulcers, genital ulcers, and skin lesions 2) oral ulcers, genital ulcers, skin lesions, and arthritis 3) oral ulcers, genital ulcers, skin lesions, and uveitis 4) oral ulcers, genital ulcers, skin lesions, and gastrointestinal involvement. A number of loci suggestive of differential risk for individual disease manifestations were proposed. Peripheral blood gene expression profile and plasma proteome exhibited significant differences in patients with different organ involvements and were able to differentiate between disease phenotypes. However, these observations require further validation and functional studies.

SUMMARY

Clustering patterns in Behçet's syndrome is highly heterogeneous. Artificial heterogeneity might obscure the true biological variation of disease expression. Preliminary genetic, transcriptomic and proteomic data suggest that different pathogenetic mechanisms may operate in different phenotypes of Behçet's syndrome.

Integration and segregation manifolds in the brain ensure cognitive flexibility during tasks and rest

Adapting to a constantly changing environment requires the human brain to flexibly switch among many demanding cognitive tasks, processing both specialized and integrated information associated with the activity in functional networks over time. In this study, we investigated the nature of the temporal alternation between segregated and integrated states in the brain during rest and six cognitive tasks using functional MRI. We employed a deep autoencoder to explore the 2D latent space associated with the segregated and integrated states. Our results show that the integrated state occupies less space in the latent space manifold compared to the segregated states. Moreover, the integrated state is characterized by lower entropy of occupancy than the segregated state, suggesting that integration plays a consolidating role, while segregation may serve as cognitive expertness. Comparing rest and the tasks, we found that rest exhibits higher entropy of occupancy, indicating a more random wandering of the mind compared to the expected focus during task performance. Our study demonstrates that both transient, short-lived integrated and segregated states are present during rest and task performance, flexibly switching between them, with integration serving as information compression and segregation related to information specialization.

Phosphoproteomic Changes Induced by Cell-Derived Matrix and Their Effect on Tumor Cell Migration and Cytoskeleton Remodeling

Increased fibrotic extracellular matrix (ECM) deposition promotes tumor invasion, which is the first step of the metastatic cascade. Yet, the underlying mechanisms are poorly understood as conventional studies of tumor cell migration are often performed in 2D cultures lacking the compositional and structural complexity of native ECM. Moreover, these studies frequently focus on select candidate pathways potentially overlooking other relevant changes in cell signaling. Here, we combine a cell-derived matrix (CDM) model with phosphotyrosine phosphoproteomic analysis to investigate tumor cell migration on fibrotic ECM relative to standard tissue culture plastic (TCP). Our results suggest that tumor cells cultured on CDMs migrate faster and in a more directional manner than their counterparts on TCP. These changes in migration correlate with decreased cell spreading and increased cell elongation. While the formation of phosphorylated focal adhesion kinase (pFAK)+ adhesion complexes did not vary between TCP and CDMs, time-dependent phosphoproteomic analysis identified that the SRC family kinase LYN may be differentially regulated. Pharmacological inhibition of LYN decreased tumor cell migration and cytoskeletal rearrangement on CDMs and also on TCP, suggesting that LYN regulates tumor cell migration on CDMs in combination with other mechanisms. These data highlight how the combination of physicochemically complex in vitro systems with phosphoproteomics can help identify signaling mechanisms by which the fibrotic ECM regulates tumor cell migration.

An unsupervised learning approach to identify immunoglobulin utilization patterns using electronic health records

BACKGROUND

Managing Canada's immunoglobulin (Ig) product resource allocation is challenging due to increasing demand, high expenditure, and global shortages. Detection of groups with high utilization rates can help with resource planning for Ig products. This study aims to uncover utilization subgroups among the Ig recipients using electronic health records (EHRs).

METHODS

The study included all Ig recipients (intravenous or subcutaneous) in Calgary from 2014 to 2020, and their EHR data, including blood inventory, recipient demographics, and laboratory test results, were analyzed. Patient clusters were derived based on patient characteristics and laboratory test data using K-means clustering. Clusters were interpreted using descriptive analyses and visualization techniques.

RESULTS

Among 4112 recipients, six clusters were identified. Clusters 1 and 2 comprised 408 (9.9%) and 1272 (30.9%) patients, respectively, contributing to 62.2% and 27.1% of total Ig utilization. Cluster 3 included 1253 (30.5%) patients, with 86.4% of infusions administered in an inpatient setting. Cluster 4, comprising 1034 (25.1%) patients, had a median age of 4 years, while clusters 2-6 were adults with median ages of 46-60. Cluster 5 had 62 (1.5%) patients, with 77.3% infusions occurring in emergency departments. Cluster 6 contained 83 (2.0%) patients receiving subcutaneous Ig treatments.

CONCLUSION

The results identified data-driven segmentations of patients with high Ig utilization rates and patients with high risk for short-term inpatient use. Our report is the first on EHR data-driven clustering of Ig utilization patterns. The findings hold the potential to inform demand forecasting and resource allocation decisions during shortages of Ig products.

KMD clustering: robust general-purpose clustering of biological data

The noisy and high-dimensional nature of biological data has spawned advanced clustering algorithms that are tailored for specific biological datatypes. However, the performance of such methods varies greatly between datasets and they require post hoc tuning of cryptic hyperparameters. We present k minimal distance (KMD) clustering, a general-purpose method based on a generalization of single and average linkage hierarchical clustering. We introduce a generalized silhouette-like function to eliminate the cryptic hyperparameter k, and use sampling to enable application to million-object datasets. Rigorous comparisons to general and specialized clustering methods on simulated, mass cytometry and scRNA-seq datasets show consistent high performance of KMD clustering across all datasets.

Single-cell transcriptomics of the ocular anterior segment: a comprehensive review

Elucidating the cellular and genetic composition of ocular tissues is essential for uncovering the pathophysiology of ocular diseases. Since the introduction of single-cell RNA sequencing (scRNA-seq) in 2009, vision researchers have performed extensive single-cell analyses to better understand transcriptome complexity and heterogeneity of ocular structures. This technology has revolutionized our ability to identify rare cell populations and to make cross-species comparisons of gene expression in both steady state and disease conditions. Importantly, single-cell transcriptomic analyses have enabled the identification of cell-type specific gene markers and signalling pathways between ocular cell populations. While most scRNA-seq studies have been conducted on retinal tissues, large-scale transcriptomic atlases pertaining to the ocular anterior segment have also been constructed in the past three years. This timely review provides vision researchers with an overview of scRNA-seq experimental design, technical limitations, and clinical applications in a variety of anterior segment-related ocular pathologies. We review open-access anterior segment-related scRNA-seq datasets and illustrate how scRNA-seq can be an indispensable tool for the development of targeted therapeutics.

Deep-Learning-Based Mixture Identification for Nuclear Magnetic Resonance Spectroscopy Applied to Plant Flavors

Nuclear magnetic resonance (NMR) is a crucial technique for analyzing mixtures consisting of small molecules, providing non-destructive, fast, reproducible, and unbiased benefits. However, it is challenging to perform mixture identification because of the offset of chemical shifts and peak overlaps that often exist in mixtures such as plant flavors. Here, we propose a deep-learning-based mixture identification method (DeepMID) that can be used to identify plant flavors (mixtures) in a formulated flavor (mixture consisting of several plant flavors) without the need to know the specific components in the plant flavors. A pseudo-Siamese convolutional neural network (pSCNN) and a spatial pyramid pooling (SPP) layer were used to solve the problems due to their high accuracy and robustness. The DeepMID model is trained, validated, and tested on an augmented data set containing 50,000 pairs of formulated and plant flavors. We demonstrate that DeepMID can achieve excellent prediction results in the augmented test set: ACC = 99.58%, TPR = 99.48%, FPR = 0.32%; and two experimentally obtained data sets: one shows ACC = 97.60%, TPR = 92.81%, FPR = 0.78% and the other shows ACC = 92.31%, TPR = 80.00%, FPR = 0.00%. In conclusion, DeepMID is a reliable method for identifying plant flavors in formulated flavors based on NMR spectroscopy, which can assist researchers in accelerating the design of flavor formulations.

Projecting genetic associations through gene expression patterns highlights disease etiology and drug mechanisms

Genes act in concert with each other in specific contexts to perform their functions. Determining how these genes influence complex traits requires a mechanistic understanding of expression regulation across different conditions. It has been shown that this insight is critical for developing new therapies. Transcriptome-wide association studies have helped uncover the role of individual genes in disease-relevant mechanisms. However, modern models of the architecture of complex traits predict that gene-gene interactions play a crucial role in disease origin and progression. Here we introduce PhenoPLIER, a computational approach that maps gene-trait associations and pharmacological perturbation data into a common latent representation for a joint analysis. This representation is based on modules of genes with similar expression patterns across the same conditions. We observe that diseases are significantly associated with gene modules expressed in relevant cell types, and our approach is accurate in predicting known drug-disease pairs and inferring mechanisms of action. Furthermore, using a CRISPR screen to analyze lipid regulation, we find that functionally important players lack associations but are prioritized in trait-associated modules by PhenoPLIER. By incorporating groups of co-expressed genes, PhenoPLIER can contextualize genetic associations and reveal potential targets missed by single-gene strategies.

Associations of functional human leucocyte antigen class I groups with HIV viral load in a heterogeneous cohort

OBJECTIVE

Human leucocyte antigen (HLA) class I alleles are the main host genetic factors involved in controlling HIV-1 viral load (VL). Nevertheless, HLA diversity has proven a significant challenge in association studies. We assessed how accounting for binding affinities of HLA class I alleles to HIV-1 peptides facilitate association testing of HLA with HIV-1 VL in a heterogeneous cohort.

DESIGN

Cohort from the Strategic Timing of AntiRetroviral Treatment (START) study.

METHODS

We imputed HLA class I alleles from host genetic data (2546 HIV+ participants) and sampled immunopeptidomes from 2079 host-paired viral genomes (targeted amplicon sequencing). We predicted HLA class I binding affinities to HIV-1 and unspecific peptides, grouping alleles into functional clusters through consensus clustering. These functional HLA class I clusters were used to test associations with HIV VL.

RESULTS

We identified four clades totaling 30 HLA alleles accounting for 11.4% variability in VL. We highlight HLA-B∗57:01 and B∗57:03 as functionally similar but yet overrepresented in distinct ethnic groups, showing when combined a protective association with HIV+ VL (log, β -0.25; adj. P-value < 0.05). We further demonstrate only a slight power reduction when using unspecific immunopeptidomes, facilitating the use of the inferred functional HLA groups in other studies.

CONCLUSION

The outlined computational approach provides a robust and efficient way to incorporate HLA function and peptide diversity, aiding clinical association studies in heterogeneous cohorts. To facilitate access to the proposed methods and results we provide an interactive application for exploring data.

Non-Specific Low Back Pain: An Inductive Exploratory Analysis through Factor Analysis and Deep Learning for Better Clustering

Non-specific low back pain (NSLBP) is a significant and pervasive public health issue in contemporary society. Despite the widespread prevalence of NSLBP, our understanding of its underlying causes, as well as our capacity to provide effective treatments, remains limited due to the high diversity in the population that does not respond to generic treatments. Clustering the NSLBP population based on shared characteristics offers a potential solution for developing personalized interventions. However, the complexity of NSLBP and the reliance on subjective categorical data in previous attempts present challenges in achieving reliable and clinically meaningful clusters. This study aims to explore the influence and importance of objective, continuous variables related to NSLBP and how to use these variables effectively to facilitate the clustering of NSLBP patients into meaningful subgroups. Data were acquired from 46 subjects who performed six simple movement tasks (back extension, back flexion, lateral trunk flexion right, lateral trunk flexion left, trunk rotation right, and trunk rotation left) at two different speeds (maximum and preferred). High-density electromyography (HD EMG) data from the lower back region were acquired, jointly with motion capture data, using passive reflective markers on the subject's body and clusters of markers on the subject's spine. An exploratory analysis was conducted using a deep neural network and factor analysis. Based on selected variables, various models were trained to classify individuals as healthy or having NSLBP in order to assess the importance of different variables. The models were trained using different subsets of data, including all variables, only anthropometric data (e.g., age, BMI, height, weight, and sex), only biomechanical data (e.g., shoulder and lower back movement), only neuromuscular data (e.g., HD EMG activity), or only balance-related data. The models achieved high accuracy in categorizing individuals as healthy or having NSLBP (full model: 93.30%, anthropometric model: 94.40%, biomechanical model: 84.47%, neuromuscular model: 88.07%, and balance model: 74.73%). Factor analysis revealed that individuals with NSLBP exhibited different movement patterns to healthy individuals, characterized by slower and more rigid movements. Anthropometric variables (age, sex, and BMI) were significantly correlated with NSLBP components. In conclusion, different data types, such as body measurements, movement patterns, and neuromuscular activity, can provide valuable information for identifying individuals with NSLBP. To gain a comprehensive understanding of NSLBP, it is crucial to investigate the main domains influencing its prognosis as a cohesive unit rather than studying them in isolation. Simplifying the conditions for acquiring dynamic data is recommended to reduce data complexity, and using back flexion and trunk rotation as effective options should be further explored.

Clustering ensemble in scRNA-seq data analysis: Methods, applications and challenges

With the rapid development of single-cell RNA-sequencing techniques, various computational methods and tools were proposed to analyze these high-throughput data, which led to an accelerated reveal of potential biological information. As one of the core steps of single-cell transcriptome data analysis, clustering plays a crucial role in identifying cell types and interpreting cellular heterogeneity. However, the results generated by different clustering methods showed distinguishing, and those unstable partitions can affect the accuracy of the analysis to a certain extent. To overcome this challenge and obtain more accurate results, currently clustering ensemble is frequently applied to cluster analysis of single-cell transcriptome datasets, and the results generated by all clustering ensembles are nearly more reliable than those from most of the single clustering partitions. In this review, we summarize applications and challenges of the clustering ensemble method in single-cell transcriptome data analysis, and provide constructive thoughts and references for researchers in this field.

Multi-Omics Profiling for Health

The world has witnessed a steady rise in both non-infectious and infectious chronic diseases, prompting a cross-disciplinary approach to understand and treating disease. Current medical care focuses on treating people after they become patients rather than preventing illness, leading to high costs in treating chronic and late-stage diseases. Additionally, a "one-size-fits all" approach to health care does not take into account individual differences in genetics, environment, or lifestyle factors, decreasing the number of people benefiting from interventions. Rapid advances in omics technologies and progress in computational capabilities have led to the development of multi-omics deep phenotyping, which profiles the interaction of multiple levels of biology over time and empowers precision health approaches. This review highlights current and emerging multi-omics modalities for precision health and discusses applications in the following areas: genetic variation, cardio-metabolic diseases, cancer, infectious diseases, organ transplantation, pregnancy, and longevity/aging. We will briefly discuss the potential of multi-omics approaches in disentangling host-microbe and host-environmental interactions. We will touch on emerging areas of electronic health record and clinical imaging integration with muti-omics for precision health. Finally, we will briefly discuss the challenges in the clinical implementation of multi-omics and its future prospects.

Spatio-temporal clustering of lumbar intervertebral flexion interactions in 127 asymptomatic individuals

The purpose of this study was to categorize asymptomatic participants based on the clustering of spatial and temporal intervertebral kinematic variables during lumbar flexion. Lumbar segmental interactions (L2-S1) were evaluated in 127 asymptomatic participants during flexion using fluoroscopy. First, four variables were identified consisting of: 1. Range of motion (ROMC), 2. Peaking time of the first derivative for separate segmentation (PTFDs), 3. Peaking magnitude of the first derivative (PMFD), and 4. Peaking time of the first derivative for stepwise (grouped) segmentation (PTFDss). These variables were used to cluster and order the lumbar levels. The number of participants required to constitute a cluster was chosen as 7. Participants formed eight (ROMC), four (PTFDs), eight (PMFD), and four (PTFDss) clusters, which included 85%, 80%, 77%, and 60% of them, respectively, according to the above features. For all clustering variables, angle time series of some lumbar levels showed significant differences between clusters. However, in general, all clusters could be categorized based on the segmental mobility contexts into three main groups as incidental macro clusters: the upper (L2-L4 > L4-S1), middle (L2-L3 < L3-L5 > L5-S1) and lower (L2-L4 < L4-S1) domains. There are spatial and temporal segmental interactions and between-subject variability in asymptomatic participants. In addition, the differences in angle time series among the clusters have provided evidence of feedback control strategies, while the stepwise segmentation facilitates consideration of the lumbar spine as a system and provides supplementary information about segmental interactions. Clinically, these facts could be taken into account when considering any intervention, but especially fusion surgery.

A biochemical necroptosis model explains cell-type-specific responses to cell death cues

Necroptosis is a form of regulated cell death associated with degenerative disorders, autoimmune and inflammatory diseases, and cancer. To better understand the biochemical mechanisms regulating necroptosis, we constructed a detailed computational model of tumor necrosis factor-induced necroptosis based on known molecular interactions from the literature. Intracellular protein levels, used as model inputs, were quantified using label-free mass spectrometry, and the model was calibrated using Bayesian parameter inference to experimental protein time course data from a well-established necroptosis-executing cell line. The calibrated model reproduced the dynamics of phosphorylated mixed lineage kinase domain-like protein, an established necroptosis reporter. A subsequent dynamical systems analysis identified four distinct modes of necroptosis signal execution, distinguished by rate constant values and the roles of the RIP1 deubiquitinating enzymes A20 and CYLD. In one case, A20 and CYLD both contribute to RIP1 deubiquitination, in another RIP1 deubiquitination is driven exclusively by CYLD, and in two modes either A20 or CYLD acts as the driver with the other enzyme, counterintuitively, inhibiting necroptosis. We also performed sensitivity analyses of initial protein concentrations and rate constants to identify potential targets for modulating necroptosis sensitivity within each mode. We conclude by associating numerous contrasting and, in some cases, counterintuitive experimental results reported in the literature with one or more of the model-predicted modes of necroptosis execution. In all, we demonstrate that a consensus pathway model of tumor necrosis factor-induced necroptosis can provide insights into unresolved controversies regarding the molecular mechanisms driving necroptosis execution in numerous cell types under different experimental conditions.

Clustering Highly Divergent Homologous Proteins: An Alignment-Free Method

The comparative analysis of amino acid sequences is an important tool in molecular biology that often requires multiple sequence alignments. In comparisons between less closely related genomes, however, it becomes more difficult to accurately align protein-coding sequences, or even to identify homologous regions in different genomes. In this article, we describe an alignment-free method for the classification of homologous protein-coding regions from different genomes. This methodology was originally developed for comparing genomes within virus families, but may be adapted for other organisms. We quantify sequence homology from the overlap (intersection distance) of the k-mer (word) frequency distributions for different protein sequences. Next, we extract groups of homologous sequences from the resulting distance matrix using a combination of dimensionality reduction and hierarchical clustering methods. Finally, we demonstrate how to generate visualizations of the composition of clusters with respect to protein annotations, and by coloring protein-coding regions of genomes by cluster assignments. These provide a useful means to quickly assess the reliability of the clustering results based on the distribution of homologous genes among genomes. © 2023 Wiley Periodicals LLC. Basic Protocol 1: Data collection and processing Basic Protocol 2: Calculating k-mer distances Basic Protocol 3: Extracting clusters of homology Support Protocol: Genome plot based on clustering results.

Sex-specific effects of calving season on joint health and biomarkers in Montana ranchers

BACKGROUND

Agricultural workers have a higher incidence of osteoarthritis (OA), but the etiology behind this phenomenon is unclear. Calving season, which occurs in mid- to late-winter for ranchers, includes physical conditions that may elevate OA risk. Our primary aim was to determine whether OA biomarkers are elevated at the peak of calving season compared to pre-season, and to compare these data with joint health survey information from the subjects. Our secondary aim was to detect biomarker differences between male and female ranchers.

METHODS

During collection periods before and during calving season, male (n = 28) and female (n = 10) ranchers completed joint health surveys and provided samples of blood, urine, and saliva for biomarker analysis. Statistical analyses examined associations between mean biomarker levels and survey predictors. Ensemble cluster analysis identified groups having unique biomarker profiles.

RESULTS

The number of calvings performed by each rancher positively correlated with plasma IL-6, serum hyaluronic acid (HA) and urinary CTX-I. Thiobarbituric acid reactive substances (TBARS), a marker of oxidative stress, was significantly higher during calving season than pre-season and was also correlated with ranchers having more months per year of joint pain. We found evidence of sexual dimorphism in the biomarkers among the ranchers, with leptin being elevated and matrix metalloproteinase-3 diminished in female ranchers. The opposite was detected in males. WOMAC score was positively associated with multiple biomarkers: IL-6, IL-2, HA, leptin, C2C, asymmetric dimethylarginine, and CTX-I. These biomarkers represent enzymatic degradation, inflammation, products of joint destruction, and OA severity.

CONCLUSIONS

The positive association between number of calvings performed by each rancher (workload) and both inflammatory and joint tissue catabolism biomarkers establishes that calving season is a risk factor for OA in Montana ranchers. Consistent with the literature, we found important sex differences in OA biomarkers, with female ranchers showing elevated leptin, whereas males showed elevated MMP-3.

Multi-omics data integration methods and their applications in psychiatric disorders

To study mental illness and health, in the past researchers have often broken down their complexity into individual subsystems (e.g., genomics, transcriptomics, proteomics, clinical data) and explored the components independently. Technological advancements and decreasing costs of high throughput sequencing has led to an unprecedented increase in data generation. Furthermore, over the years it has become increasingly clear that these subsystems do not act in isolation but instead interact with each other to drive mental illness and health. Consequently, individual subsystems are now analysed jointly to promote a holistic understanding of the underlying biological complexity of health and disease. Complementing the increasing data availability, current research is geared towards developing novel methods that can efficiently combine the information rich multi-omics data to discover biologically meaningful biomarkers for diagnosis, treatment, and prognosis. However, clinical translation of the research is still challenging. In this review, we summarise conventional and state-of-the-art statistical and machine learning approaches for discovery of biomarker, diagnosis, as well as outcome and treatment response prediction through integrating multi-omics and clinical data. In addition, we describe the role of biological model systems and in silico multi-omics model designs in clinical translation of psychiatric research from bench to bedside. Finally, we discuss the current challenges and explore the application of multi-omics integration in future psychiatric research. The review provides a structured overview and latest updates in the field of multi-omics in psychiatry.

GeoWaVe: geometric median clustering with weighted voting for ensemble clustering of cytometry data

MOTIVATION

Clustering is an unsupervised method for identifying structure in unlabelled data. In the context of cytometry, it is typically used to categorize cells into subpopulations of similar phenotypes. However, clustering is greatly dependent on hyperparameters and the data to which it is applied as each algorithm makes different assumptions and generates a different 'view' of the dataset. As such, the choice of clustering algorithm can significantly influence results, and there is often not one preferred method but different insights to be obtained from different methods. To overcome these limitations, consensus approaches are needed that directly address the effect of competing algorithms. To the best of our knowledge, consensus clustering algorithms designed specifically for the analysis of cytometry data are lacking.

RESULTS

We present a novel ensemble clustering methodology based on geometric median clustering with weighted voting (GeoWaVe). Compared to graph ensemble clustering methods that have gained popularity in single-cell RNA sequencing analysis, GeoWaVe performed favourably on different sets of high-dimensional mass and flow cytometry data. Our findings provide proof of concept for the power of consensus methods to make the analysis, visualization and interpretation of cytometry data more robust and reproducible. The wide availability of ensemble clustering methods is likely to have a profound impact on our understanding of cellular responses, clinical conditions and therapeutic and diagnostic options.

AVAILABILITY AND IMPLEMENTATION

GeoWaVe is available as part of the CytoCluster package https://github.com/burtonrj/CytoCluster and published on the Python Package Index https://pypi.org/project/cytocluster. Benchmarking data described are available from https://doi.org/10.5281/zenodo.7134723.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Subchondral bone structure and synovial fluid metabolism are altered in injured and contralateral limbs 7 days after non-invasive joint injury in skeletally-mature C57BL/6 mice

OBJECTIVE

Post-traumatic osteoarthritis (PTOA) commonly develops after ACL injury, but early changes to the joint soon after injury are insufficiently understood. The objectives of this study were (1) evaluate the response of subchondral bone tissue modulus to joint injury and (2) identify which bone structural, material, and metabolic outcomes are local (i.e., injured joint only) or systemic (i.e., injured and contralateral-to-injured).

DESIGN

Female C57Bl∖6N mice (19 weeks at injury) underwent tibial compression overload to simulate ACL injury (n = 8) or a small pre-load (n = 8). Synovial fluid was harvested at euthanasia 7 days later for metabolomic profiling. Bone outcomes included epiphyseal and SCB microarchitecture, SCB nanoindentation modulus, SCB formation rate, and osteoclast number density.

RESULTS

Injury decreased epiphyseal bone volume fraction ([-5.29, -1.38%], P = 0.0016) and decreased SCB thickness for injured vs sham-injured limbs ([2.2, 31.4 μm], P = 0.017)). Epiphyseal bone loss commonly occurred for contralateral-to-injured limbs. There was not sufficient evidence to conclude that SCB modulus changes with injury. Metabolomic analyses revealed dysregulated synovial fluid metabolism with joint injury but that many metabolic pathways are shared between injured and contralateral-to-injured limbs.

CONCLUSION

This study demonstrates rapid changes to bone structure and synovial fluid metabolism after injury with the potential for influencing the progression to PTOA. These changes are often evidenced in the contralateral-to-injured limb, indicating that systemic musculoskeletal responses to joint injury should not be overlooked.

Emerging Computational Methods in Mass Spectrometry Imaging

Mass spectrometry imaging (MSI) is a powerful analytical technique that generates maps of hundreds of molecules in biological samples with high sensitivity and molecular specificity. Advanced MSI platforms with capability of high-spatial resolution and high-throughput acquisition generate vast amount of data, which necessitates the development of computational tools for MSI data analysis. In addition, computation-driven MSI experiments have recently emerged as enabling technologies for further improving the MSI capabilities with little or no hardware modification. This review provides a critical summary of computational methods and resources developed for MSI data analysis and interpretation along with computational approaches for improving throughput and molecular coverage in MSI experiments. This review is focused on the recently developed artificial intelligence methods and provides an outlook for a future paradigm shift in MSI with transformative computational methods.

Further validation of the Cognitive Biases Questionnaire for psychosis

BACKGROUND

Cognitive biases are recognized as important treatment targets for reducing symptoms associated with severe mental disorders. Although cognitive biases have been linked to symptoms in most studies, few studies have looked at such biases transdiagnostically. The Cognitive Bias Questionnaire for psychosis (CBQp) is a self-reported questionnaire that assesses cognitive biases amongst individuals with a psychotic disorder, as well as individuals with other severe mental disorders. The current study aims to validate a French version of the CBQp and to explore transdiagnostic cognitive biases in individuals with psychotic disorders, individuals with depression, and in healthy controls.

METHODS

The CBQp was translated into French following a protocol based on international standards. Discriminant validity and internal consistency were determined for total score and each subscale score. Confirmatory factor analyses were performed to test construct validity. Finally, cluster analyses were conducted to investigate cognitive biases across diagnostic groups.

RESULTS

Our results were similar to those of the original authors, with the one-factor solution (assessment of a general thinking bias) being the strongest, but the two-factor solution (assessing biases within two themes relating to psychosis) and the five-factor solution (assessment of multiple distinct biases) being clinically more interesting. A six-cluster solution emerged, suggesting that individuals with similar diagnoses score differently on all cognitive biases, and that individuals with different diagnoses might have similar cognitive biases.

CONCLUSIONS

The current findings support the validity of the French translation of the CBQp. Our cluster analyses overall support the transdiagnostic presence of cognitive biases.

Multi-omics assessment of dilated cardiomyopathy using non-negative matrix factorization

Dilated cardiomyopathy (DCM), a myocardial disease, is heterogeneous and often results in heart failure and sudden cardiac death. Unavailability of cardiac tissue has hindered the comprehensive exploration of gene regulatory networks and nodal players in DCM. In this study, we carried out integrated analysis of transcriptome and methylome data using non-negative matrix factorization from a cohort of DCM patients to uncover underlying latent factors and covarying features between whole-transcriptome and epigenome omics datasets from tissue biopsies of living patients. DNA methylation data from Infinium HM450 and mRNA Illumina sequencing of n = 33 DCM and n = 24 control probands were filtered, analyzed and used as input for matrix factorization using R NMF package. Mann-Whitney U test showed 4 out of 5 latent factors are significantly different between DCM and control probands (P<0.05). Characterization of top 10% features driving each latent factor showed a significant enrichment of biological processes known to be involved in DCM pathogenesis, including immune response (P = 3.97E-21), nucleic acid binding (P = 1.42E-18), extracellular matrix (P = 9.23E-14) and myofibrillar structure (P = 8.46E-12). Correlation network analysis revealed interaction of important sarcomeric genes like Nebulin, Tropomyosin alpha-3 and ERC-protein 2 with CpG methylation of ATPase Phospholipid Transporting 11A0, Solute Carrier Family 12 Member 7 and Leucine Rich Repeat Containing 14B, all with significant P values associated with correlation coefficients >0.7. Using matrix factorization, multi-omics data derived from human tissue samples can be integrated and novel interactions can be identified. Hypothesis generating nature of such analysis could help to better understand the pathophysiology of complex traits such as DCM.

Trusting our machines: validating machine learning models for single-molecule transport experiments

In this tutorial review, we will describe crucial aspects related to the application of machine learning to help users avoid the most common pitfalls. The examples we present will be based on data from the field of molecular electronics, specifically single-molecule electron transport experiments, but the concepts and problems we explore will be sufficiently general for application in other fields with similar data. In the first part of the tutorial review, we will introduce the field of single-molecule transport, and provide an overview of the most common machine learning algorithms employed. In the second part of the tutorial review, we will show, through examples grounded in single-molecule transport, that the promises of machine learning can only be fulfilled by careful application. We will end the tutorial review with a discussion of where we, as a field, could go from here.

Broad Whitefish (Coregonus nasus) isotopic niches: Stable isotopes reveal diverse foraging strategies and habitat use in Arctic Alaska

Understanding the ecological niche of some fishes is complicated by their frequent use of a broad range of food resources and habitats across space and time. Little is known about Broad Whitefish (Coregonus nasus) ecological niches in Arctic landscapes even though they are an important subsistence species for Alaska’s Indigenous communities. We investigated the foraging ecology and habitat use of Broad Whitefish via stable isotope analyses of muscle and liver tissue and otoliths from mature fish migrating in the Colville River within Arctic Alaska. The range of δ¹³C (-31.8– -21.9‰) and δ¹⁵N (6.6–13.1‰) across tissue types and among individuals overlapped with isotope values previously observed in Arctic lakes and rivers, estuaries, and nearshore marine habitat. The large range of δ¹⁸O (4.5–10.9‰) and δD (-237.6– -158.9‰) suggests fish utilized a broad spectrum of habitats across elevational and latitudinal gradients. Cluster analysis of muscle δ¹³Cˈ, δ¹⁵N, δ¹⁸O, and δD indicated that Broad Whitefish occupied four different foraging niches that relied on marine and land-based (i.e., freshwater and terrestrial) food sources to varying degrees. Most individuals had isotopic signatures representative of coastal freshwater habitat (Group 3; 25%) or coastal lagoon and delta habitat (Group 1; 57%), while individuals that mainly utilized inland freshwater (Group 4; 4%) and nearshore marine habitats (Group 2; 14%) represented smaller proportions. Otolith microchemistry confirmed that individuals with more enriched muscle tissue δ¹³Cˈ, δD, and δ¹⁸O tended to use marine habitats, while individuals that mainly used freshwater habitats had values that were less enriched. The isotopic niches identified here represent important foraging habitats utilized by Broad Whitefish. To preserve access to these diverse habitats it will be important to limit barriers along nearshore areas and reduce impacts like roads and climate change on natural flow regimes. Maintaining these diverse connected habitats will facilitate long-term population stability, buffering populations from future environmental and anthropogenic perturbations.

Modes of cognition: Evidence from metastable brain dynamics

Managing cognitive load depends on adequate resource allocation by the human brain through the engagement of metastable substates, which are large-scale functional networks that change over time. We employed a novel analysis method, deep autoencoder dynamical analysis (DADA), with 100 healthy adults selected from the Human Connectome Project (HCP) data set in rest and six cognitive tasks. The deep autoencoder of DADA described seven recurrent stochastic metastable substates from the functional connectome of BOLD phase coherence matrices. These substates were significantly differentiated in terms of their probability of appearance, time duration, and spatial attributes. We found that during different cognitive tasks, there was a higher probability of having more connected substates dominated by a high degree of connectivity in the thalamus. In addition, compared with those during tasks, resting brain dynamics have a lower level of predictability, indicating a more uniform distribution of metastability between substates, quantified by higher entropy. These novel findings provide empirical evidence for the philosophically motivated cognitive theory, suggesting on-line and off-line as two fundamentally distinct modes of cognition. On-line cognition refers to task-dependent engagement with the sensory input, while off-line cognition is a slower, environmentally detached mode engaged with decision and planning. Overall, the DADA framework provides a bridge between neuroscience and cognitive theory that can be further explored in the future.

Statistical power for cluster analysis

BACKGROUND

Cluster algorithms are gaining in popularity in biomedical research due to their compelling ability to identify discrete subgroups in data, and their increasing accessibility in mainstream software. While guidelines exist for algorithm selection and outcome evaluation, there are no firmly established ways of computing a priori statistical power for cluster analysis. Here, we estimated power and classification accuracy for common analysis pipelines through simulation. We systematically varied subgroup size, number, separation (effect size), and covariance structure. We then subjected generated datasets to dimensionality reduction approaches (none, multi-dimensional scaling, or uniform manifold approximation and projection) and cluster algorithms (k-means, agglomerative hierarchical clustering with Ward or average linkage and Euclidean or cosine distance, HDBSCAN). Finally, we directly compared the statistical power of discrete (k-means), "fuzzy" (c-means), and finite mixture modelling approaches (which include latent class analysis and latent profile analysis).

RESULTS

We found that clustering outcomes were driven by large effect sizes or the accumulation of many smaller effects across features, and were mostly unaffected by differences in covariance structure. Sufficient statistical power was achieved with relatively small samples (N = 20 per subgroup), provided cluster separation is large (Δ = 4). Finally, we demonstrated that fuzzy clustering can provide a more parsimonious and powerful alternative for identifying separable multivariate normal distributions, particularly those with slightly lower centroid separation (Δ = 3).

CONCLUSIONS

Traditional intuitions about statistical power only partially apply to cluster analysis: increasing the number of participants above a sufficient sample size did not improve power, but effect size was crucial. Notably, for the popular dimensionality reduction and clustering algorithms tested here, power was only satisfactory for relatively large effect sizes (clear separation between subgroups). Fuzzy clustering provided higher power in multivariate normal distributions. Overall, we recommend that researchers (1) only apply cluster analysis when large subgroup separation is expected, (2) aim for sample sizes of N = 20 to N = 30 per expected subgroup, (3) use multi-dimensional scaling to improve cluster separation, and (4) use fuzzy clustering or mixture modelling approaches that are more powerful and more parsimonious with partially overlapping multivariate normal distributions.

Using networks to analyze and visualize the distribution of overlapping genes in virus genomes

Gene overlap occurs when two or more genes are encoded by the same nucleotides. This phenomenon is found in all taxonomic domains, but is particularly common in viruses, where it may increase the information content of compact genomes or influence the creation of new genes. Here we report a global comparative study of overlapping open reading frames (OvRFs) of 12,609 virus reference genomes in the NCBI database. We retrieved metadata associated with all annotated open reading frames (ORFs) in each genome record to calculate the number, length, and frameshift of OvRFs. Our results show that while the number of OvRFs increases with genome length, they tend to be shorter in longer genomes. The majority of overlaps involve +2 frameshifts, predominantly found in dsDNA viruses. Antisense overlaps in which one of the ORFs was encoded in the same frame on the opposite strand (−0) tend to be longer. Next, we develop a new graph-based representation of the distribution of overlaps among the ORFs of genomes in a given virus family. In the absence of an unambiguous partition of ORFs by homology at this taxonomic level, we used an alignment-free k-mer based approach to cluster protein coding sequences by similarity. We connect these clusters with two types of directed edges to indicate (1) that constituent ORFs are adjacent in one or more genomes, and (2) that these ORFs overlap. These adjacency graphs not only provide a natural visualization scheme, but also a novel statistical framework for analyzing the effects of gene- and genome-level attributes on the frequencies of overlaps.

Multi-viewpoints visual models for efficient modeling and analysis of Twitter based health-care services

Purpose

The purpose of this study for eHealth text mining domains, cosine-based visual methods (VM) assess the clusters more accurately than Euclidean; which are recommended for tweet data models for clusters assessment. Such VM determines the clusters concerning a single viewpoint or none, which are less informative. Multi-viewpoints (MVP) were used for addressing the more informative clusters assessment of health-care tweet documents and to demonstrate visual analysis of cluster tendency.

Design/methodology/approach

In this paper, the authors proposed MVP-based VM by using traditional topic models with visual techniques to find cluster tendency, partitioning for cluster validity to propose health-care recommendations based on tweets. The authors demonstrated the effectiveness of proposed methods on different real-time Twitter health-care data sets in the experimental study. The authors also did a comparative analysis of proposed models with existing visual assessment tendency (VAT) and cVAT models by using cluster validity indices and computational complexities; the examples suggest that MVP VM were more informative.

Findings

In this paper, the authors proposed MVP-based VM by using traditional topic models with visual techniques to find cluster tendency, partitioning for cluster validity to propose health-care recommendations based on tweets.

Originality/value

In this paper, the authors proposed multi-viewpoints distance metric in topic model cluster tendency for the first time and visual representation using VAT images using hybrid topic models to find cluster tendency, partitioning for cluster validity to propose health-care recommendations based on tweets.

Unsupervised quantification of naturalistic animal behaviors for gaining insight into the brain

Neural computation has evolved to optimize the behaviors that enable our survival. Although much previous work in neuroscience has focused on constrained task behaviors, recent advances in computer vision are fueling a trend toward the study of naturalistic behaviors. Automated tracking of fine-scale behaviors is generating rich datasets for animal models including rodents, fruit flies, zebrafish, and worms. However, extracting meaning from these large and complex data often requires sophisticated computational techniques. Here we review the latest methods and modeling approaches providing new insights into the brain from behavior. We focus on unsupervised methods for identifying stereotyped behaviors and for resolving details of the structure and dynamics of behavioral sequences.

A Constrained Feature Selection Approach Based on Feature Clustering and Hypothesis Margin Maximization

In this paper, we propose a semisupervised feature selection approach that is based on feature clustering and hypothesis margin maximization. The aim is to improve the classification accuracy by choosing the right feature subset and to allow building more interpretable models. Our approach handles the two core aspects of feature selection, i.e., relevance and redundancy, and is divided into three steps. First, the similarity weights between features are represented by a sparse graph where each feature can be reconstructed from the sparse linear combination of the others. Second, features are then hierarchically clustered identifying groups of the most similar ones. Finally, a semisupervised margin-based objective function is optimized to select the most data discriminative feature from within each cluster, hence maximizing relevance while minimizing redundancy among features. Eventually, we empirically validate our proposed approach on multiple well-known UCI benchmark datasets in terms of classification accuracy and representation entropy, where it proved to outperform four other semisupervised and unsupervised methods and competed with two widely used supervised ones.

Automated and unbiased discrimination of ALS from control tissue at single cell resolution

Histopathological analysis of tissue sections is invaluable in neurodegeneration research. However, cell-to-cell variation in both the presence and severity of a given phenotype is a key limitation of this approach, reducing the signal to noise ratio and leaving unresolved the potential of single-cell scoring for a given disease attribute. Here, we tested different machine learning methods to analyse high-content microscopy measurements of hundreds of motor neurons (MNs) from amyotrophic lateral sclerosis (ALS) post-mortem tissue sections. Furthermore, we automated the identification of phenotypically distinct MN subpopulations in VCP- and SOD1-mutant transgenic mice, revealing common morphological cellular phenotypes. Additionally we established scoring metrics to rank cells and tissue samples for both disease probability and severity. By adapting this paradigm to human post-mortem tissue, we validated our core finding that morphological descriptors robustly discriminate ALS from control healthy tissue at single cell resolution. Determining disease presence, severity and unbiased phenotypes at single cell resolution might prove transformational in our understanding of ALS and neurodegeneration more broadly.

Identification of Clinically Relevant Subgroups of Chronic Lymphocytic Leukemia Through Discovery of Abnormal Molecular Pathways

Chronic lymphocytic leukemia (CLL) is the most common form of adult leukemia in the Western world with a highly variable clinical course. Its striking genetic heterogeneity is not yet fully understood. Although the CLL genetic landscape has been well-described, patient stratification based on mutation profiles remains elusive mainly due to the heterogeneity of data. Here we attempted to decrease the heterogeneity of somatic mutation data by mapping mutated genes in the respective biological processes. From the sequencing data gathered by the International Cancer Genome Consortium for 506 CLL patients, we generated pathway mutation scores, applied ensemble clustering on them, and extracted abnormal molecular pathways with a machine learning approach. We identified four clusters differing in pathway mutational profiles and time to first treatment. Interestingly, common CLL drivers such as ATM or TP53 were associated with particular subtypes, while others like NOTCH1 or SF3B1 were not. This study provides an important step in understanding mutational patterns in CLL.

Intricate Genetic Programs Controlling Dormancy in Mycobacterium tuberculosis

Mycobacterium tuberculosis (MTB) displays the remarkable ability to transition in and out of dormancy, a hallmark of the pathogen’s capacity to evade the immune system and exploit susceptible individuals. Uncovering the gene regulatory programs that underlie the phenotypic shifts in MTB during disease latency and reactivation has posed a challenge. We develop an experimental system to precisely control dissolved oxygen levels in MTB cultures in order to capture the transcriptional events that unfold as MTB transitions into and out of hypoxia-induced dormancy. Using a comprehensive genome-wide transcription factor binding map and insights from network topology analysis, we identify regulatory circuits that deterministically drive sequential transitions across six transcriptionally and functionally distinct states encompassing more than three-fifths of the MTB genome. The architecture of the genetic programs explains the transcriptional dynamics underlying synchronous entry of cells into a dormant state that is primed to infect the host upon encountering favorable conditions.

Mycobacterium tuberculosis (MTB) persists within the host by counteracting disparate stressors including hypoxia. Peterson et al. report a transcriptional program that coordinates sequential state transitions to drive MTB in and out of hypoxia-induced dormancy. Among varied properties, this program encodes advanced preparedness to infect the host in favorable conditions.

Denoising large-scale biological data using network filters

Background

Large-scale biological data sets are often contaminated by noise, which can impede accurate inferences about underlying processes. Such measurement noise can arise from endogenous biological factors like cell cycle and life history variation, and from exogenous technical factors like sample preparation and instrument variation.

Results

We describe a general method for automatically reducing noise in large-scale biological data sets. This method uses an interaction network to identify groups of correlated or anti-correlated measurements that can be combined or “filtered” to better recover an underlying biological signal. Similar to the process of denoising an image, a single network filter may be applied to an entire system, or the system may be first decomposed into distinct modules and a different filter applied to each. Applied to synthetic data with known network structure and signal, network filters accurately reduce noise across a wide range of noise levels and structures. Applied to a machine learning task of predicting changes in human protein expression in healthy and cancerous tissues, network filtering prior to training increases accuracy up to 43% compared to using unfiltered data.

Conclusions

Network filters are a general way to denoise biological data and can account for both correlation and anti-correlation between different measurements. Furthermore, we find that partitioning a network prior to filtering can significantly reduce errors in networks with heterogenous data and correlation patterns, and this approach outperforms existing diffusion based methods. Our results on proteomics data indicate the broad potential utility of network filters to applications in systems biology.

Supplementary Information

The online version supplementary material available at 10.1186/s12859-021-04075-x.

Spatial Segmentation of Mass Spectrometry Imaging Data by Combining Multivariate Clustering and Univariate Thresholding

Spatial segmentation partitions mass spectrometry imaging (MSI) data into distinct regions, providing a concise visualization of the vast amount of data and identifying regions of interest (ROIs) for downstream statistical analysis. Unsupervised approaches are particularly attractive, as they may be used to discover the underlying subpopulations present in the high-dimensional MSI data without prior knowledge of the properties of the sample. Herein, we introduce an unsupervised spatial segmentation approach, which combines multivariate clustering and univariate thresholding to generate comprehensive spatial segmentation maps of the MSI data. This approach combines matrix factorization and manifold learning to enable high-quality image segmentation without an extensive hyperparameter search. In parallel, some ion images inadequately represented in the multivariate analysis were treated using univariate thresholding to generate complementary spatial segments. The final spatial segmentation map was assembled from segment candidates that were generated using both techniques. We demonstrate the performance and robustness of this approach for two MSI data sets of mouse uterine and kidney tissue sections that were acquired with different spatial resolutions. The resulting segmentation maps are easy to interpret and project onto the known anatomical regions of the tissue.

Integration of AI and traditional medicine in drug discovery

AI integration in plant-based traditional medicine could be used to overcome drug discovery challenges.

Co-stimulation with opposing macrophage polarization cues leads to orthogonal secretion programs in individual cells

Macrophages are innate immune cells that contribute to fighting infections, tissue repair, and maintaining tissue homeostasis. To enable such functional diversity, macrophages resolve potentially conflicting cues in the microenvironment via mechanisms that are unclear. Here, we use single-cell RNA sequencing to explore how individual macrophages respond when co-stimulated with inflammatory stimuli LPS and IFN-γ and the resolving cytokine IL-4. These co-stimulated macrophages display a distinct global transcriptional program. However, variable negative cross-regulation between some LPS + IFN-γ-specific and IL-4-specific genes results in cell-to-cell heterogeneity in transcription. Interestingly, negative cross-regulation leads to mutually exclusive expression of the T-cell-polarizing cytokine genes Il6 and Il12b versus the IL-4-associated factors Arg1 and Chil3 in single co-stimulated macrophages, and single-cell secretion measurements show that these specialized functions are maintained for at least 48 h. This study suggests that increasing functional diversity in the population is one strategy macrophages use to respond to conflicting environmental cues.