transcriptomics

MIRCID is a framework that infers hub-miRNAs to enhance drug mechanism-of-action modeling by comparing gene expression with inferred transcription factor activity and miRNA expression, achieving improvements in pathway classification and similarity-based retrieval.

BioM-JEPA introduces a joint-embedding predictive architecture that learns single-cell representations by predicting graph-connected gene blocks instead of individual genes, showing improved efficiency and downstream performance in perturbation-response tasks.

This paper introduces MS-MLB, an open machine learning benchmark for classifying multiple sclerosis from whole blood RNA expression data using the public GSE17048 cohort. It provides a reproducible, leakage-controlled evaluation pipeline and reports Gradient Boosting as the top performer.

CASCADE is an agentic framework that predicts downstream transcriptional effects of gene perturbations using regulatory networks, with patient-data validation across TCGA cancer types. The paper demonstrates directional concordance for MYC and other regulators, though it does not exceed existing curated knowledge.

This paper introduces CoCoS, a contrastive pretraining framework that learns whole-cell representations from complementary transcriptomic views, addressing limitations of masked gene reconstruction in single-cell foundation models. Experiments on cell-type annotation and gene regulatory network inference show competitive transfer performance.

This paper systematically benchmarks classical machine learning models (Random Forest, XGBoost, etc.) for ER status prediction using multi-omics data from TCGA-BRCA, finding that RNA expression provides the strongest predictive signal and that Random Forest achieves 90.3% balanced accuracy in the integrated multi-omic setting.

PerturbCellRL introduces a reinforcement learning framework that post-trains a pretrained single-cell transcriptomic generator using cell-level verifiers as rewards, improving biological consistency of perturbation predictions beyond distributional matching.

A new preprint from the Arc Institute introduces PerturbSpace, a method for spatially resolved, multimodal whole-transcriptome CRISPR screens compatible with standard single-cell workflows.

This paper formalizes transcriptome-based drug design (TBDD) as a generative inverse problem and proposes CURE, a multi-resolution transcriptome-guided diffusion framework that generates drug molecules conditioned on desired transcriptomic state transitions.