R.ROSETTA: an interpretable machine learning framework

Using machine learning methods to study the tumour microenvironment and its biomarkers in osteosarcoma metastasis

Background

The long-term prognosis for patients with osteosarcoma (OS) metastasis remains unfavourable, highlighting the urgent need for research that explores potential biomarkers using innovative methodologies.

Methods

This study explored potential biomarkers for OS metastasis by analysing data from the Cancer Genome Atlas Program (TCGA) and Gene Expression Omnibus (GEO) databases. The synthetic minority oversampling technique (SMOTE) was employed to tackle class imbalances, while genes were selected using four feature selection algorithms (Monte Carlo feature selection [MCFS], Borota, minimum-redundancy maximum-relevance [mRMR], and light gradient-boosting machine [LightGBM]) based on the gene expression matrix. Four machine learning (ML) algorithms (support vector machine [SVM], extreme gradient boosting [XGBoost], random forest [RF], and k-nearest neighbours [kNN]) were utilized to determine the optimal number of genes for building the model. Interpretable machine learning (IML) was applied to construct prediction networks, revealing potential relationships among the selected genes. Additionally, enrichment analysis, survival analysis, and immune infiltration were performed on the featured genes.

Results

In DS1, DS2, and DS3, the IML algorithm identified 53, 45, and 46 features, respectively. Using the merged gene set, we obtained a total of 79 interpretable prediction rules for OS metastasis. We subsequently conducted an in-depth investigation on 39 crucial molecules associated with predicting OS metastasis, elucidating their roles within the tumour microenvironment. Importantly, we found that certain genes act as both predictors and differentially expressed genes. Finally, our study unveiled statistically significant differences in survival between the high and low expression groups of TRIP4, S100A9, SELL and SLC11A1, and there was a certain correlation between these genes and 22 various immune cells.

Conclusions

The biomarkers discovered in this study hold significant implications for personalized therapies, potentially enhancing the clinical prognosis of patients with OS.

The role of artificial neural networks in prediction of severe acute pancreatitis associated acute respiratory distress syndrome: A retrospective study

Early identification and intervention of acute respiratory distress syndrome (ARDS) are particularly important. This study aimed to construct predictive models for ARDS following severe acute pancreatitis (SAP) by artificial neural networks and logistic regression. The artificial neural networks model was constructed using clinical data from 214 SAP patients. The patient cohort was randomly divided into a training set and a test set, with 149 patients allocated to the training set and 65 patients assigned to the test set. The artificial neural networks and logistic regression models were trained by the training set, and then the performance of both models was evaluated using the test set. The sensitivity, specificity, PPV, NPV, accuracy, and AUC value of artificial neural networks model were 68.0%, 87.5%, 77.3%, 81.4%, 80.0%, 0.853 ± 0.054 (95% CI: 0.749-0.958). The sensitivity, specificity, PPV, NPV, accuracy and AUC value of logistic regression model were 48.7%, 85.3%, 65.5%, 74.4%, 72.0%, 0.799 ± 0.045 (95% CI: 0.710-0.888). There were no significant differences between the artificial neural networks and logistic regression models in predictive performance. Bedside Index of Severity in Acute Pancreatitis score, procalcitonin, prothrombin time, and serum calcium were the most important predictive variables in the artificial neural networks model. The discrimination abilities of logistic regression and artificial neural networks models in predicting SAP-related ARDS were similar. It is advisable to choose the model according to the specific research purpose.

Using Machine Learning Methods to Study Colorectal Cancer Tumor Micro-Environment and Its Biomarkers

Colorectal cancer (CRC) is a leading cause of cancer deaths worldwide, and the identification of biomarkers can improve early detection and personalized treatment. In this study, RNA-seq data and gene chip data from TCGA and GEO were used to explore potential biomarkers for CRC. The SMOTE method was used to address class imbalance, and four feature selection algorithms (MCFS, Borota, mRMR, and LightGBM) were used to select genes from the gene expression matrix. Four machine learning algorithms (SVM, XGBoost, RF, and kNN) were then employed to obtain the optimal number of genes for model construction. Through interpretable machine learning (IML), co-predictive networks were generated to identify rules and uncover underlying relationships among the selected genes. Survival analysis revealed that INHBA, FNBP1, PDE9A, HIST1H2BG, and CADM3 were significantly correlated with prognosis in CRC patients. In addition, the CIBERSORT algorithm was used to investigate the proportion of immune cells in CRC tissues, and gene mutation rates for the five selected biomarkers were explored. The biomarkers identified in this study have significant implications for the development of personalized therapies and could ultimately lead to improved clinical outcomes for CRC patients.

Applications of Machine Learning in Knowledge Management System: A Comprehensive Review

As new generations of technology appear, legacy knowledge management solutions and applications become increasingly out of date, necessitating a paradigm shift. Machine learning presents an opportunity by foregoing rule-based knowledge intensive systems inundating the marketplace. An extensive review was made on the literature pertaining to machine learning which common machine learning algorithms were identified. This study has analysed more than 200 papers extracted from Scopus and IEEE databases. Searches ranged with the bulk of the articles from 2018 to 2021, while some articles ranged from 1959 to 2017. The research gap focusses on implementing machine learning algorithm to knowledge management systems, specifically knowledge management attributes. By investigating and reviewing each algorithm extensively, the usability of each algorithm is identified, with its advantages and disadvantages. From there onwards, these algorithms were mapped for what area of knowledge management it may be beneficial. Based on the findings, it is evidently seen how these algorithms are applicable in knowledge management and how it can enhance knowledge management system further. Based on the findings, the paper aims to bridge the gap between the literature in knowledge management and machine learning. A knowledge management–machine learning framework is conceived based on the review done on each algorithm earlier and to bridge the gap between the two literatures. The framework highlights how machine learning algorithm can play a part in different areas of knowledge management. From the framework, it provides practitioners how and where to implement machine learning in knowledge management.

Interpretable machine learning identifies paediatric Systemic Lupus Erythematosus subtypes based on gene expression data

Transcriptomic analyses are commonly used to identify differentially expressed genes between patients and controls, or within individuals across disease courses. These methods, whilst effective, cannot encompass the combinatorial effects of genes driving disease. We applied rule-based machine learning (RBML) models and rule networks (RN) to an existing paediatric Systemic Lupus Erythematosus (SLE) blood expression dataset, with the goal of developing gene networks to separate low and high disease activity (DA1 and DA3). The resultant model had an 81% accuracy to distinguish between DA1 and DA3, with unsupervised hierarchical clustering revealing additional subgroups indicative of the immune axis involved or state of disease flare. These subgroups correlated with clinical variables, suggesting that the gene sets identified may further the understanding of gene networks that act in concert to drive disease progression. This included roles for genes (i) induced by interferons (IFI35 and OTOF), (ii) key to SLE cell types (KLRB1 encoding CD161), or (iii) with roles in autophagy and NF-κB pathway responses (CKAP4). As demonstrated here, RBML approaches have the potential to reveal novel gene patterns from within a heterogeneous disease, facilitating patient clinical and therapeutic stratification.

Machine Learning-Based Analysis of Glioma Grades Reveals Co-Enrichment

Gliomas develop and grow in the brain and central nervous system. Examining glioma grading processes is valuable for improving therapeutic challenges. One of the most extensive repositories storing transcriptomics data for gliomas is The Cancer Genome Atlas (TCGA). However, such big cohorts should be processed with caution and evaluated thoroughly as they can contain batch and other effects. Furthermore, biological mechanisms of cancer contain interactions among biomarkers. Thus, we applied an interpretable machine learning approach to discover such relationships. This type of transparent learning provides not only good predictability, but also reveals co-predictive mechanisms among features. In this study, we corrected the strong and confounded batch effect in the TCGA glioma data. We further used the corrected datasets to perform comprehensive machine learning analysis applied on single-sample gene set enrichment scores using collections from the Molecular Signature Database. Furthermore, using rule-based classifiers, we displayed networks of co-enrichment related to glioma grades. Moreover, we validated our results using the external glioma cohorts. We believe that utilizing corrected glioma cohorts from TCGA may improve the application and validation of any future studies. Finally, the co-enrichment and survival analysis provided detailed explanations for glioma progression and consequently, it should support the targeted treatment.

A Review of Interpretable ML in Healthcare: Taxonomy, Applications, Challenges, and Future Directions

We have witnessed the impact of ML in disease diagnosis, image recognition and classification, and many more related fields. Healthcare is a sensitive field related to people’s lives in which decisions need to be carefully taken based on solid evidence. However, most ML models are complex, i.e., black-box, meaning they do not provide insights into how the problems are solved or why such decisions are proposed. This lack of interpretability is the main reason why some ML models are not widely used yet in real environments such as healthcare. Therefore, it would be beneficial if ML models could provide explanations allowing physicians to make data-driven decisions that lead to higher quality service. Recently, several efforts have been made in proposing interpretable machine learning models to become more convenient and applicable in real environments. This paper aims to provide a comprehensive survey and symmetry phenomena of IML models and their applications in healthcare. The fundamental characteristics, theoretical underpinnings needed to develop IML, and taxonomy for IML are presented. Several examples of how they are applied in healthcare are investigated to encourage and facilitate the use of IML models in healthcare. Furthermore, current limitations, challenges, and future directions that might impact applying ML in healthcare are addressed.

Machine learning predictive models for acute pancreatitis: A systematic review

INTRODUCTION

Acute pancreatitis (AP) is a common clinical pancreatic disease. Patients with different severity levels have different clinical outcomes. With the advantages of algorithms, machine learning (ML) has gradually emerged in the field of disease prediction, assisting doctors in decision-making.

METHODS

A systematic review was conducted using the PubMed, Web of Science, Scopus, and Embase databases, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Publication time was limited from inception to 29 May 2021. Studies that have used ML to establish predictive tools for AP were eligible for inclusion. Quality assessment of the included studies was conducted in accordance with the IJMEDI checklist.

RESULTS

In this systematic review, 24 of 2,913 articles, with a total of 8,327 patients and 47 models, were included. The studies could be divided into five categories: 10 studies (42%) reported severity prediction; 10 studies (42%), complication prediction; 3 studies (13%), mortality prediction; 2 studies (8%), recurrence prediction; and 2 studies (8%), surgery timing prediction. ML showed great accuracy in several prediction tasks. However, most of the included studies were retrospective in nature, conducted at a single centre, based on database data, and lacked external validation. According to the IJMEDI checklist and our scoring criteria, two studies were considered to be of high quality. Most studies had an obvious bias in the quality of data preparation, validation, and deployment dimensions.

CONCLUSION

In the prediction tasks for AP, ML has shown great potential in assisting decision-making. However, the existing studies still have some deficiencies in the process of model construction. Future studies need to optimize the deficiencies and further evaluate the comparability of the ML systems and model performance, so as to consequently develop high-quality ML-based models that can be used in clinical practice.

Transcriptomic analysis reveals pro-inflammatory signatures associated with acute myeloid leukemia progression

Progression of AML is associated with pro-inflammatory mediators through altered expression levels of CR1, DPEP1, IL1R1, and ST18.

Upregulated CD6 and downregulated INSR are nodes in gene expression networks linked to AML relapse, according to machine learning analysis.

Numerous studies have been performed over the last decade to exploit the complexity of genomic and transcriptomic lesions driving the initiation of acute myeloid leukemia (AML). These studies have helped improve risk classification and treatment options. Detailed molecular characterization of longitudinal AML samples is sparse, however; meanwhile, relapse and therapy resistance represent the main challenges in AML care. To this end, we performed transcriptome-wide RNA sequencing of longitudinal diagnosis, relapse, and/or primary resistant samples from 47 adult and 23 pediatric AML patients with known mutational background. Gene expression analysis revealed the association of short event-free survival with overexpression of GLI2 and IL1R1, as well as downregulation of ST18. Moreover, CR1 downregulation and DPEP1 upregulation were associated with AML relapse both in adults and children. Finally, machine learning–based and network-based analysis identified overexpressed CD6 and downregulated INSR as highly copredictive genes depicting important relapse-associated characteristics among adult patients with AML. Our findings highlight the importance of a tumor-promoting inflammatory environment in leukemia progression, as indicated by several of the herein identified differentially expressed genes. Together, this knowledge provides the foundation for novel personalized drug targets and has the potential to maximize the benefit of current treatments to improve cure rates in AML.

Interpretable Machine Learning Reveals Dissimilarities Between Subtypes of Autism Spectrum Disorder

Autism spectrum disorder (ASD) is a heterogeneous neuropsychiatric disorder with a complex genetic background. Analysis of altered molecular processes in ASD patients requires linear and nonlinear methods that provide interpretable solutions. Interpretable machine learning provides legible models that allow explaining biological mechanisms and support analysis of clinical subgroups. In this work, we investigated several case-control studies of gene expression measurements of ASD individuals. We constructed a rule-based learning model from three independent datasets that we further visualized as a nonlinear gene-gene co-predictive network. To find dissimilarities between ASD subtypes, we scrutinized a topological structure of the network and estimated a centrality distance. Our analysis revealed that autism is the most severe subtype of ASD, while pervasive developmental disorder-not otherwise specified and Asperger syndrome are closely related and milder ASD subtypes. Furthermore, we analyzed the most important ASD-related features that were described in terms of gene co-predictors. Among others, we found a strong co-predictive mechanism between EMC4 and TMEM30A, which may suggest a co-regulation between these genes. The present study demonstrates the potential of applying interpretable machine learning in bioinformatics analyses. Although the proposed methodology was designed for transcriptomics data, it can be applied to other omics disciplines.