Integration of language model and geometric deep learning enables accurate and efficient genome-scale annotation of comprehensive protein-ligand binding sites.

The Mouse Action Recognition System is a computational pipeline for automated classification of social behaviors in freely interacting mice, accompanied by a graphical interface for analysis of multimodal neuroscience datasets.

A computational method identifies the functions of orphan enzymes by organizing them into metabolic pathways; the prediction of a new l-gulonate catabolic pathway is experimentally tested and confirmed.

GENESPACE lets users track related chromosomal sequences across multiple reference genomes.

Computational efforts yield a zebrafish transcriptome annotation that provides more sensitive and comprehensive gene detection over existing annotations when used for bulk or single-cell RNA-seq.

A near-complete flux balance analysis model of a minimal cell demonstrates the high essentiality of its metabolic genes, agrees well with experimental essentiality data and suggests some further gene removals.

Evidence for those pathogen genes selected specifically for success on crops.

Computational models and software connect metagenomics to metabolic network reconstruction, assess metabolic complementarity between species, and identify critical species associated to functions of interest.

Genome neighborhood networks provide an efficient large-scale approach to mine metabolic pathway context for entire families of proteins and enzymes.

One hundred one high-quality drosophilid genomes are released, along with low-cost assembly workflows, as an open community resource for studying genetics, ecology, and evolution in this important model system.