T. M. Murali

This paper introduces CoEvoLink, a computational framework that predicts virus-host interactions by integrating sequence data with coevolutionary phylogenetic models, improving prediction accuracy over existing methods.

HAVEN predicts viral hosts from protein sequences, achieving high accuracy for both common and rare hosts while outperforming larger models.

NEFFy computes the Number of Effective Sequences (NEFF) for multiple sequence alignments, offering integrated functionality and support for diverse formats and sequence types.

SynVerse evaluates DL models for drug synergy prediction, finding none outperform a naive baseline and highlighting poor generalization to unseen drugs and cell lines.

ICoN integrates multiple protein-protein association networks using co-attention and denoising to improve downstream tasks like gene module detection and protein function prediction.

The paper introduces Ensemble Integration (EI), a late integration method for multimodal biomedical data, achieving more accurate predictions than individual modalities and early integration approaches.

The paper builds a computational model of the nutrient signaling network in budding yeast to reconcile experimental data with a proposed molecular mechanism and test predictions in mutant strains.

The paper develops FastSinkSource, a label propagation algorithm for gene function prediction in bacterial networks, achieving high accuracy and efficiency across 200 species.

The paper quantitatively estimates genetic interactions between 36 yeast cell-cycle genes using high-throughput phenotyping of 7,350 mutants and refines a mathematical cell cycle model.

This study benchmarks GRN inference algorithms using synthetic and Boolean models, finding moderate accuracy and highlighting challenges in ensemble methods.

The paper evaluates connectivity in signaling pathways using different graph representations and introduces a new measure to quantify molecular and pathway influence.

The paper develops algorithms to efficiently compute genetic crosses for creating target mutants from source mutants, considering verifiability and permissibility constraints.