Biology subjects

Pershad, Y.

Publications and source records attributed to Pershad, Y..

3 recordsLinked to original sources

bioRxiv2025

Closing the loop: Teaching single-cell foundation models to learn from perturbations

The application of transfer learning models to large scale single-cell datasets has enabled the development of single-cell foundation models (scFMs) that can predict cellular responses to perturbations in silico. Although these predictions can be experimentally tested, current scFMs are unable to "close the loop" and learn from these experiments to create better predictions. Here, we introduce a "closed-loop" framework that extends the scFM by incorporating perturbation data during model fine-tuning. Our closed-loop model improves prediction accuracy, increasing positive predictive value in the setting of T-cell activation three-fold. We applied this model to RUNX1-familial platelet disorder, a rare pediatric blood disorder and identified two therapeutic targets (mTOR and CD74-MIF signaling axis) and two novel pathways (protein kinase C and phosphoinositide 3-kinase). This work establishes that iterative incorporation of experimental data to foundation models enhances biological predictions, representing a crucial step toward realizing the promise of "virtual cell" models for biomedical discovery.

bioinformatics↗

bioRxiv2025

Inherited resilience to clonal hematopoiesis by modifying stem cell RNA regulation

Somatic mutations that increase hematopoietic stem cell (HSC) fitness drive their expansion in clonal hematopoiesis (CH) and predispose to blood cancers. Although CH frequently occurs with aging, it rarely progresses to overt malignancy. Population variation in the growth rate and potential of mutant clones suggests the presence of genetic factors protecting against CH, but these remain largely undefined. Here, we identify a non-coding regulatory variant, rs17834140-T, that significantly protects against CH and myeloid malignancies by downregulating HSC-selective expression and function of the RNA-binding protein MSI2. By modeling variant effects and mapping MSI2 binding targets, we uncover an RNA network that maintains human HSCs and influences CH risk. Importantly, rs17834140-T is associated with slower CH expansion rates in humans, and stem cell MSI2 levels modify ASXL1-mutant HSC clonal dominance in experimental models. These findings leverage natural resilience to highlight a key role for post-transcriptional regulation in human HSCs, and offer genetic evidence supporting inhibition of MSI2 or its downstream targets as rational strategies for blood cancer prevention.

genetics↗

bioRxiv2019

Pathway and network embedding methods for prioritizing psychiatric drugs

One in five Americans experience mental illness, and roughly 75% of psychiatric prescriptions do not successfully treat the patients condition. Extensive evidence implicates genetic factors and signaling disruption in the pathophysiology of these diseases. Changes in transcription often underlie this molecular pathway dysregulation; individual patient transcriptional data can improve the efficacy of diagnosis and treatment. Recent large-scale genomic studies have uncovered shared genetic modules across multiple psychiatric disorders--providing an opportunity for an integrated multi-disease approach for diagnosis. Moreover, network-based models informed by gene expression can represent pathological biological mechanisms and suggest new genes for diagnosis and treatment. Here, we use patient gene expression data from multiple studies to classify psychiatric diseases, integrate knowledge from expert-curated databases and publicly available experimental data to create augmented disease-specific gene sets, and use these to recommend disease-relevant drugs. From Gene Expression Omnibus, we extract expression data from 145 cases of schizophrenia, 82 cases of bipolar disorder, 190 cases of major depressive disorder, and 307 shared controls. We use pathway-based approaches to predict psychiatric disease diagnosis with a random forest model (78% accuracy) and derive important features to augment available drug and disease signatures. Using protein-protein-interaction networks and embedding-based methods, we build a pipeline to prioritize treatments for psychiatric diseases that achieves a 3.4-fold improvement over a background model. Thus, we demonstrate that gene-expression-derived pathway features can diagnose psychiatric diseases and that molecular insights derived from this classification task can inform treatment prioritization for psychiatric diseases.

bioinformatics↗