Dileep Kishore

This study shows that filtering genomes based on coherent genotype-phenotype relationships improves cross-dataset prediction of microbial carbon utilisation, outperforming larger but less coherent datasets.

This paper develops a machine learning framework using genome annotations to predict bacterial phenotypes, achieving high accuracy for Gram-stain type and respiration type.

The paper introduces MIND, a platform for comparing and integrating microbial interaction networks from diverse datasets.

The paper introduces cayenne, a Python package for stochastic simulations, and benchmarks its accuracy against other libraries, finding it offers the best balance of accuracy and speed.

Pipeliner provides a Nextflow-based framework for creating flexible and reproducible sequencing data processing pipelines, supporting bulk RNA-seq, scRNA-seq, and DGE data.