Abdur Rahman M. A. Basher

This paper introduces PHet, a method for selecting features that preserve disease heterogeneity and improve subtype discovery in single-cell data.

The paper introduces reMap, a framework for relabeling metabolic pathway data into groups to enhance pathway prediction performance across genome complexity levels.

This paper introduces triUMPF, a triple non-negative matrix factorization method for metabolic pathway inference, achieving improved precision on multi-organismal datasets.

The paper introduces pathway2vec, a software package that uses network embedding to predict metabolic pathways from genomic data, showing improved pathway prediction outcomes.

The paper introduces mlLGPR, a multi-label logistic regression tool for metabolic pathway prediction using rich pathway features, and benchmarks its performance across diverse datasets.

leADS improves metabolic pathway inference by using active dataset subsampling to address class imbalance in training data.