Biology subjects

Lussier, Y. A.

Publications and source records attributed to Lussier, Y. A..

5 recordsLinked to original sources

bioRxiv2025

Paired-Sample and Pathway-Anchored MLOps Framework for Robust Transcriptomic Machine Learning in Small Cohorts

BackgroundNinety percent of the 65,000 human diseases are infrequent, collectively affecting [~] 400 million people, substantially limiting cohort accrual. This low prevalence constrains the development of robust transcriptome-based machine learning (ML) classifiers. Standard data-driven classifiers typically require cohorts of over 100 subjects per group to achieve clinical accuracy while managing high-dimensional input ([~]25,000 transcripts). These requirements are infeasible for micro-cohorts of [~]20 individuals, where overfitting becomes pervasive. ObjectiveTo overcome these constraints, we developed a classification method that integrates three enabling strategies: (i) paired-sample transcriptome dynamics, (ii) N-of-1 pathway-based analytics, and (iii) reproducible machine learning operations (MLOps) for continuous model refinement. MethodsUnlike ML approaches relying on a single transcriptome per subject, within-subject paired-sample designs -- such as pre-versus post-treatment or diseased versus adjacent-normal tissue -- effectively control intra-individual variability under isogenic conditions and within-subject environmental exposures (e.g. smoking history, other medications, etc.), improve signal-to-noise ratios, and, when pre-processed as single-subject studies (N-of-1), can achieve statistical power comparable to that obtained in animal models. Pathway-level N-of-1 analytics further reduces each samples high-dimensional profile into [~]4,000 biologically interpretable features, annotated with effect sizes, dispersion, and significance. Complementary MLOps practices--automated versioning, continuous monitoring, and adaptive hyperparameter tuning--improve model reproducibility and generalization. ResultsIn two case studies--human rhinovirus infection versus matched healthy controls (n=16 training; 3 test) and breast cancer tissues harboring TP53 or PIK3CA mutations versus adjacent normal tissue (n=27 training; 9 test)--this approach achieved 90% precision and recall on an unseen breast cancer test set and 92% precision with 90% recall in rhinovirus fivefold cross-validation. Incorporating paired-sample dynamics boosted precision by 8.8% and recall by 6%, while the MLOps workflow yielded additional gains of 14.5% and 12.5%, respectively. Moreover, our method identified 42 critical gene sets (pathways) for rhinovirus response and 21 for cancer mutation status. ConclusionsThese proof-of-concept results support the utility of integrating intra-subject dynamics, "biological knowledge"-based feature reduction (pathway-level feature reduction grounded in prior biological knowledge; e.g., N-of-1-pathways analytics), and reproducible MLOps workflows can overcome cohort-size limitations in infrequent disease, offering a scalable, interpretable solution for high-dimensional transcriptomic classification. Future work will extend these advances across various therapeutic and small-cohort designs.

bioinformatics↗

bioRxiv2021

"Single-subject studies"-derived analyses unveil altered biomechanisms between very small cohorts: implications for rare diseases

MotivationIdentifying altered transcripts between very small human cohorts is particularly challenging and is compounded by the low accrual rate of human subjects in rare diseases or sub-stratified common disorders. Yet, single-subject studies (S3) can compare paired transcriptome samples drawn from the same patient under two conditions (e.g., treated vs pre-treatment) and suggest patient-specific responsive biomechanisms based on the overrepresentation of functionally defined gene sets. These improve statistical power by: (i) reducing the total features tested and (ii) relaxing the requirement of within-cohort uniformity at the transcript level. We propose Inter-N-of-1, a novel method, to identify meaningful biomechanism differences between very small cohorts by using the effect size of "single-subject-study"-derived responsive biomechanisms. ResultsIn each subject, Inter-N-of-1 requires applying previously published S3-type N-of-1-pathways MixEnrich to two paired samples (e.g., diseased vs unaffected tissues) for determining patient-specific enriched genes sets: Odds Ratios (S3-OR) and S3-variance using Gene Ontology Biological Processes. To evaluate small cohorts, we calculated the precision and recall of Inter-N-of-1 and that of a control method (GLM+EGS) when comparing two cohorts of decreasing sizes (from 20 vs 20 to 2 vs 2) in a comprehensive six-parameter simulation and in a proof-of-concept clinical dataset. In simulations, the Inter-N-of-1 median precision and recall are > 90% and >75% in cohorts of 3 vs 3 distinct subjects (regardless of the parameter values), whereas conventional methods outperform Inter-N-of-1 at sample sizes 9 vs 9 and larger. Similar results were obtained in the clinical proof-of-concept dataset. AvailabilityR software is available at Lussierlab.net/BSSD. ContactLussier.y@gmail.com, Piegorsch@math.arizona.edu

bioinformatics↗

bioRxiv2019

binomialRF: Scalable Feature Selection and Screening for Random Forests to Identify Biomarkers and Their Interactions

BackgroundIn this era of data science-driven bioinformatics, machine learning research has focused on feature selection as users want more interpretation and post-hoc analyses for biomarker detection. However, when there are more features (i.e., transcript) than samples (i.e., mice or human samples) in a study, this poses major statistical challenges in biomarker detection tasks as traditional statistical techniques are underpowered in high dimension. Second and third order interactions of these features pose a substantial combinatoric dimensional challenge. In computational biology, random forest1 (RF) classifiers are widely used2-7 due to their flexibility, powerful performance, and robustness to "P predictors >> subjects N" difficulties and their ability to rank features. We propose binomialRF, a feature selection technique in RFs that provides an alternative interpretation for features using a correlated binomial distribution and scales efficiently to analyze multiway interactions. MethodsbinomialRF treats each tree in a RF as a correlated but exchangeable binary trial. It determines importance by constructing a test statistic based on a features selection frequency to compute its rank, nominal p-value, and multiplicity-adjusted q-value using a one-sided hypothesis test with a correlated binomial distribution. A distributional adjustment addresses the co-dependencies among trees as these trees subsample from the same dataset. The proposed algorithm efficiently identifies multiway nonlinear interactions by generalizing the test statistic to count sub-trees. ResultsIn simulations and in the Madelon benchmark datasets studies, binomialRF showed computational gains (up to 30 to 600 times faster) while maintaining competitive variable precision and recall in identifying biomarkers main effects and interactions. In two clinical studies, the binomialRF algorithm prioritizes previously-published relevant pathological molecular mechanisms (features) with high classification precision and recall using features alone, as well as with their statistical interactions alone. ConclusionbinomialRF extends upon previous methods for identifying interpretable features in RFs and brings them together under a correlated binomial distribution to create an efficient hypothesis testing algorithm that identifies biomarkers main effects and interactions. Preliminary results in simulations demonstrate computational gains while retaining competitive model selection and classification accuracies. Future work will extend this framework to incorporate ontologies that provide path-way-level feature selection from gene expression input data. AvailabilityGithub: https://github.com/SamirRachidZaim/binomialRF Supplementary informationSupplementary analyses and results are available at https://github.com/SamirRachidZaim/binomialRF_simulationStudy

bioinformatics↗

bioRxiv2018

Evaluating single-subject study methods for personal transcriptomic interpretations to advance precision medicine

BackgroundGene expression profiling has benefited medicine by providing clinically relevant insights at the molecular candidate and systems levels. However, to adopt a more precision approach that integrates individual variability including omics data into risk assessments, diagnoses, and therapeutic decision making, whole transcriptome expression analysis requires methodological advancements. One need is for users to confidently be able to make individual-level inferences from whole transcriptome data. We propose that biological replicates in isogenic conditions can provide a framework for testing differentially expressed genes (DEGs) in a single subject (ss) in absence of an appropriate external reference standard or replicates.\n\nMethodsEight ss methods for identifying genes with differential expression (NOISeq, DEGseq, edgeR, mixture model, DESeq, DESeq2, iDEG, and ensemble) were compared in Yeast (parental line versus snf2 deletion mutant; n=42/condition) and MCF7 breast-cancer cell (baseline and stimulated with estradiol; n=7/condition) RNA-Seq datasets where replicate analysis was used to build reference standards from NOISeq, DEGseq, edgeR, DESeq, DESeq2. Each dataset was randomly partitioned so that approximately two-thirds of the paired samples were used to construct reference standards and the remainder were treated separately as single-subject sample pairs and DEGs were assayed using ss methods. Receiver-operator characteristic (ROC) and precision-recall plots were determined for all ss methods against each RSs in both datasets (525 combinations).\n\nResultsConsistent with prior analyses of these data, ~50% and ~15% DEGs were respectively obtained in Yeast and MCF7 reference standard datasets regardless of the analytical method. NOISeq, edgeR and DESeq were the most concordant and robust methods for creating a reference standard. Single-subject versions of NOISeq, DEGseq, and an ensemble learner achieved the best median ROC-area-under-the-curve to compare two transcriptomes without replicates regardless of the type of reference standard (>90% in Yeast, >0.75 in MCF7).\n\nConclusionBetter and more consistent accuracies are obtained by an ensemble method applied to singlesubject studies across different conditions. In addition, distinct specific sing-subject methods perform better according to different proportions of DEGs. Single-subject methods for identifying DEGs from paired samples need improvement, as no method performs with both precision>90% and recall>90%. http://www.lussiergroup.org/publications/EnsembleBiomarker

bioinformatics↗

bioRxiv2018

iDEG: A single-subject method for assessing gene differential expression from two transcriptomes of an individual

BackgroundAccurate profiling of gene expression in a single subject has the potential to be a powerful precision medicine tool, useful for unveiling individual disease mechanisms and responses. However, expression analysis tools for RNA-sequencing (RNA-Seq) data require replicate samples to estimate gene-wise data variability and make inferences, which is costly and not easily obtainable in clinical practice. Strategies to implement DEGSeq, DESeq, and edgeR for comparing two conditions without replicates (TCWR) have been proposed without evaluation, while NOISeq-sim was validated in a restricted way using qPCR on 400 transcripts. These methods impose restrictive assumptions in TCWR limiting inferential opportunities.\n\nMethodsWe propose a new method that borrows information across different genes from the same individual using a partitioned window to strategically bypass the requirement of replicates per condition. We termed this method \"iDEG\", which identifies individualized Differentially Expressed Genes in a single subject sampled under two conditions without replicates, i.e., a baseline sample (unaffected tissue) vs. a case sample (tumor). iDEG transforms RNA-Seq data such that, under the null hypothesis, differences of transformed expression counts follow a distribution and variance calculated across a local partition of related transcripts at baseline expression. This transformation enables modeling genes with a two-group mixture model from which the probability of differential expression for each gene is then estimated by an empirical Bayes approach with a local false discovery rate control. To compare the performance of iDEG to other methods applied to TCWR, we conducted simulations assuming a Negative Binomial distribution with varying dispersion parameters and percentages of differentially expressed genes (DEGs).\n\nResultsOur extensive simulation studies demonstrate that iDEGs F1 accuracy scores better than the other methods at 5% 90% and recall>75% and low false positive rate (<1%) in most conditions.\n\nConclusionThe partitioned window strategy provides a novel and accurate way to borrow information across genes locally and would probably increase the accuracy of all relevant methods.

bioinformatics↗