A primer on machine learning techniques for genomic applications

Machine Learning Approach and Bioinformatics Analysis Discovered Key Genomic Signatures for Hepatitis B Virus-Associated Hepatocyte Remodeling and Hepatocellular Carcinoma

Hepatitis B virus (HBV) causes liver cancer, which is the third most common cause of cancer-related death worldwide. Chronic inflammation via HBV in the host hepatocytes causes hepatocyte remodeling (hepatocyte transformation and immortalization) and hepatocellular carcinoma (HCC). Recognizing cancer stages accurately to optimize early screening and diagnosis is a primary concern in the outlook of HBV-induced hepatocyte remodeling and liver cancer. Genomic signatures play important roles in addressing this issue. Recently, machine learning (ML) models and bioinformatics analysis have become very important in discovering novel genomic signatures for the early diagnosis, treatment, and prognosis of HBV-induced hepatic cell remodeling and HCC. We discuss the recent literature on the ML approach and bioinformatics analysis revealed novel genomic signatures for diagnosing and forecasting HBV-associated hepatocyte remodeling and HCC. Various genomic signatures, including various microRNAs and their associated genes, long noncoding RNAs (lncRNAs), and small nucleolar RNAs (snoRNAs), have been discovered to be involved in the upregulation and downregulation of HBV-HCC. Moreover, these genetic biomarkers also affect different biological processes, such as proliferation, migration, circulation, assault, dissemination, antiapoptosis, mitogenesis, transformation, and angiogenesis in HBV-infected hepatocytes.

Missing genotype imputation in non-model species using self-organizing maps

Current methodologies of genome-wide single-nucleotide polymorphism (SNP) genotyping produce large amounts of missing data that may affect statistical inference and bias the outcome of experiments. Genotype imputation is routinely used in well-studied species to buffer the impact in downstream analysis, and several algorithms are available to fill in missing genotypes. The lack of reference haplotype panels precludes the use of these methods in genomic studies on non-model organisms. As an alternative, machine learning algorithms are employed to explore the genotype data and to estimate the missing genotypes. Here, we propose an imputation method based on self-organizing maps (SOM), a widely used neural networks formed by spatially distributed neurons that cluster similar inputs into close neurons. The method explores genotype datasets to select SNP loci to build binary vectors from the genotypes, and initializes and trains neural networks for each query missing SNP genotype. The SOM-derived clustering is then used to impute the best genotype. To automate the imputation process, we have implemented gtImputation, an open-source application programmed in Python3 and with a user-friendly GUI to facilitate the whole process. The method performance was validated by comparing its accuracy, precision and sensitivity on several benchmark genotype datasets with other available imputation algorithms. Our approach produced highly accurate and precise genotype imputations even for SNPs with alleles at low frequency and outperformed other algorithms, especially for datasets from mixed populations with unrelated individuals.

Genome language modeling (GLM): a beginner's cheat sheet

Integrating genomics with diverse data modalities has the potential to revolutionize personalized medicine. However, this integration poses significant challenges due to the fundamental differences in data types and structures. The vast size of the genome necessitates transformation into a condensed representation containing key biomarkers and relevant features to ensure interoperability with other modalities. This commentary explores both conventional and state-of-the-art approaches to genome language modeling (GLM), with a focus on representing and extracting meaningful features from genomic sequences. We focus on the latest trends of applying language modeling techniques on genomics sequence data, treating it as a text modality. Effective feature extraction is essential in enabling machine learning models to effectively analyze large genomic datasets, particularly within multimodal frameworks. We first provide a step-by-step guide to various genomic sequence preprocessing and tokenization techniques. Then we explore feature extraction methods for the transformation of tokens using frequency, embedding, and neural network-based approaches. In the end, we discuss machine learning (ML) applications in genomics, focusing on classification, regression, language processing algorithms, and multimodal integration. Additionally, we explore the role of GLM in functional annotation, emphasizing how advanced ML models, such as Bidirectional encoder representations from transformers, enhance the interpretation of genomic data. To the best of our knowledge, we compile the first end-to-end analytic guide to convert complex genomic data into biologically interpretable information using GLM, thereby facilitating the development of novel data-driven hypotheses.

REDInet: a temporal convolutional network-based classifier for A-to-I RNA editing detection harnessing million known events

A-to-I ribonucleic acid (RNA) editing detection is still a challenging task. Current bioinformatics tools rely on empirical filters and whole genome sequencing or whole exome sequencing data to remove background noise, sequencing errors, and artifacts. Sometimes they make use of cumbersome and time-consuming computational procedures. Here, we present REDInet, a temporal convolutional network-based deep learning algorithm, to profile RNA editing in human RNA sequencing (RNAseq) data. It has been trained on REDIportal RNA editing sites, the largest collection of human A-to-I changes from >8000 RNAseq data of the genotype-tissue expression project. REDInet can classify editing events with high accuracy harnessing RNAseq nucleotide frequencies of 101-base windows without the need for coupled genomic data.

When less is more: sketching with minimizers in genomics

The exponential increase in sequencing data calls for conceptual and computational advances to extract useful biological insights. One such advance, minimizers, allows for reducing the quantity of data handled while maintaining some of its key properties. We provide a basic introduction to minimizers, cover recent methodological developments, and review the diverse applications of minimizers to analyze genomic data, including de novo genome assembly, metagenomics, read alignment, read correction, and pangenomes. We also touch on alternative data sketching techniques including universal hitting sets, syncmers, or strobemers. Minimizers and their alternatives have rapidly become indispensable tools for handling vast amounts of data.

Where developmental toxicity meets explainable artificial intelligence: state-of-the-art and perspectives

INTRODUCTION

The application of Artificial Intelligence (AI) to predictive toxicology is rapidly increasing, particularly aiming to develop non-testing methods that effectively address ethical concerns and reduce economic costs. In this context, Developmental Toxicity (Dev Tox) stands as a key human health endpoint, especially significant for safeguarding maternal and child well-being.

AREAS COVERED

This review outlines the existing methods employed in Dev Tox predictions and underscores the benefits of utilizing New Approach Methodologies (NAMs), specifically focusing on eXplainable Artificial Intelligence (XAI), which proves highly efficient in constructing reliable and transparent models aligned with recommendations from international regulatory bodies.

EXPERT OPINION

The limited availability of high-quality data and the absence of dependable Dev Tox methodologies render XAI an appealing avenue for systematically developing interpretable and transparent models, which hold immense potential for both scientific evaluations and regulatory decision-making.

Explainable artificial intelligence and microbiome data for food geographical origin: the Mozzarella di Bufala Campana PDO Case of Study

Identifying the origin of a food product holds paramount importance in ensuring food safety, quality, and authenticity. Knowing where a food item comes from provides crucial information about its production methods, handling practices, and potential exposure to contaminants. Machine learning techniques play a pivotal role in this process by enabling the analysis of complex data sets to uncover patterns and associations that can reveal the geographical source of a food item. This study aims to investigate the potential use of explainable artificial intelligence for identifying the food origin. The case of study of Mozzarella di Bufala Campana PDO has been considered by examining the composition of the microbiota in each samples. Three different supervised machine learning algorithms have been compared and the best classifier model is represented by Random Forest with an Area Under the Curve (AUC) value of 0.93 and the top accuracy of 0.87. Machine learning models effectively classify origin, offering innovative ways to authenticate regional products and support local economies. Further research can explore microbiota analysis and extend applicability to diverse food products and contexts for enhanced accuracy and broader impact.

Predictive Value of Machine Learning for Platinum Chemotherapy Responses in Ovarian Cancer: Systematic Review and Meta-Analysis

BACKGROUND

Machine learning is a potentially effective method for predicting the response to platinum-based treatment for ovarian cancer. However, the predictive performance of various machine learning methods and variables is still a matter of controversy and debate.

OBJECTIVE

This study aims to systematically review relevant literature on the predictive value of machine learning for platinum-based chemotherapy responses in patients with ovarian cancer.

METHODS

Following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, we systematically searched the PubMed, Embase, Web of Science, and Cochrane databases for relevant studies on predictive models for platinum-based therapies for the treatment of ovarian cancer published before April 26, 2023. The Prediction Model Risk of Bias Assessment tool was used to evaluate the risk of bias in the included articles. Concordance index (C-index), sensitivity, and specificity were used to evaluate the performance of the prediction models to investigate the predictive value of machine learning for platinum chemotherapy responses in patients with ovarian cancer.

RESULTS

A total of 1749 articles were examined, and 19 of them involving 39 models were eligible for this study. The most commonly used modeling methods were logistic regression (16/39, 41%), Extreme Gradient Boosting (4/39, 10%), and support vector machine (4/39, 10%). The training cohort reported C-index in 39 predictive models, with a pooled value of 0.806; the validation cohort reported C-index in 12 predictive models, with a pooled value of 0.831. Support vector machine performed well in both the training and validation cohorts, with a C-index of 0.942 and 0.879, respectively. The pooled sensitivity was 0.890, and the pooled specificity was 0.790 in the training cohort.

CONCLUSIONS

Machine learning can effectively predict how patients with ovarian cancer respond to platinum-based chemotherapy and may provide a reference for the development or updating of subsequent scoring systems.

Microbiome preterm birth DREAM challenge: Crowdsourcing machine learning approaches to advance preterm birth research

Every year, 11% of infants are born preterm with significant health consequences, with the vaginal microbiome a risk factor for preterm birth. We crowdsource models to predict (1) preterm birth (PTB; <37 weeks) or (2) early preterm birth (ePTB; <32 weeks) from 9 vaginal microbiome studies representing 3,578 samples from 1,268 pregnant individuals, aggregated from public raw data via phylogenetic harmonization. The predictive models are validated on two independent unpublished datasets representing 331 samples from 148 pregnant individuals. The top-performing models (among 148 and 121 submissions from 318 teams) achieve area under the receiver operator characteristic (AUROC) curve scores of 0.69 and 0.87 predicting PTB and ePTB, respectively. Alpha diversity, VALENCIA community state types, and composition are important features in the top-performing models, most of which are tree-based methods. This work is a model for translation of microbiome data into clinically relevant predictive models and to better understand preterm birth.

Deep learning in bioinformatics

Deep learning is a powerful machine learning technique that can learn from large amounts of data using multiple layers of artificial neural networks. This paper reviews some applications of deep learning in bioinformatics, a field that deals with analyzing and interpreting biological data. We first introduce the basic concepts of deep learning and then survey the recent advances and challenges of applying deep learning to various bioinformatics problems, such as genome sequencing, gene expression analysis, protein structure prediction, drug discovery, and disease diagnosis. We also discuss future directions and opportunities for deep learning in bioinformatics. We aim to provide an overview of deep learning so that bioinformaticians applying deep learning models can consider all critical technical and ethical aspects. Thus, our target audience is biomedical informatics researchers who use deep learning models for inference. This review will inspire more bioinformatics researchers to adopt deep-learning methods for their research questions while considering fairness, potential biases, explainability, and accountability.

Microbiome Preterm Birth DREAM Challenge: Crowdsourcing Machine Learning Approaches to Advance Preterm Birth Research

Globally, every year about 11% of infants are born preterm, defined as a birth prior to 37 weeks of gestation, with significant and lingering health consequences. Multiple studies have related the vaginal microbiome to preterm birth. We present a crowdsourcing approach to predict: (a) preterm or (b) early preterm birth from 9 publicly available vaginal microbiome studies representing 3,578 samples from 1,268 pregnant individuals, aggregated from raw sequences via an open-source tool, MaLiAmPi. We validated the crowdsourced models on novel datasets representing 331 samples from 148 pregnant individuals. From 318 DREAM challenge participants we received 148 and 121 submissions for our two separate prediction sub-challenges with top-ranking submissions achieving bootstrapped AUROC scores of 0.69 and 0.87, respectively. Alpha diversity, VALENCIA community state types, and composition (via phylotype relative abundance) were important features in the top performing models, most of which were tree based methods. This work serves as the foundation for subsequent efforts to translate predictive tests into clinical practice, and to better understand and prevent preterm birth.

A Machine Learning Approach to Parkinson’s Disease Blood Transcriptomics

The increased incidence and the significant health burden associated with Parkinson’s disease (PD) have stimulated substantial research efforts towards the identification of effective treatments and diagnostic procedures. Despite technological advancements, a cure is still not available and PD is often diagnosed a long time after onset when irreversible damage has already occurred. Blood transcriptomics represents a potentially disruptive technology for the early diagnosis of PD. We used transcriptome data from the PPMI study, a large cohort study with early PD subjects and age matched controls (HC), to perform the classification of PD vs. HC in around 550 samples. Using a nested feature selection procedure based on Random Forests and XGBoost we reached an AUC of 72% and found 493 candidate genes. We further discussed the importance of the selected genes through a functional analysis based on GOs and KEGG pathways.