Nature’s computational biology and bioinformatics collection presents a snapshot of how data driven methods and artificial intelligence are reshaping life sciences and adjacent fields. Recent news, features and research articles span topics from psychotherapy style tests on chatbots to projects that map human genome organization, reflecting the breadth of disciplines now intertwined with computational approaches. Alongside core biological work on cis regulatory elements, chromatin structure and microbiomes, the page foregrounds debates about artificial intelligence safety, regulation and its growing role in research, education and the economy.
Several primary research papers focus on decoding gene regulation and cellular behavior using large scale datasets and advanced modelling. An expanded registry of candidate human and mouse cis regulatory elements builds on previous ENCODE resources by integrating new functional data and additional cell and tissue types. Bidirectional CRISPR screens paired with single cell and spatial analysis identify the transcription factor GLIS3 as a driver of chronic inflammation and fibrosis and as a potential marker of ulcerative colitis severity. The 4D Nucleome Project combines genomic assays with computational methods to predict genome folding and to infer how genetic variants, including disease associated variants, influence genome structure and function. Other studies map the architecture of the neutrophil compartment across health, inflammation and cancer, and dissect immune responses in a pig to human kidney xenotransplant through dense longitudinal multi omics profiling across the 61-day procedure.
Computational design and generative models feature prominently in protein and gene engineering. One open access article describes a generative artificial intelligence powered method for de novo design of highly active metallohydrolases using only the geometry of active site residues, without requiring backbone or sequence information. A complementary strategy that combines machine learning and atomistic modelling enables one shot design of efficient enzymes for diverse biological and non biological transformations. A genomic language model called Evo learns a semantics of gene function based on genomic context so that it can autocomplete DNA prompts to generate de novo genes encoding proteins and RNAs with defined activities that generalize beyond natural sequences. Large scale microbiome work, including the Microflora Danica atlas of Danish environmental microbiomes and multi national gut microbiome studies, links microbial diversity and species recurrence to environmental disturbance and to cardiometabolic health markers, while other content explores pancreatic cancer models and the polyclonal origins of premalignant colorectal lesions.
Across news, podcasts and opinion, the collection tracks how artificial intelligence permeates scientific practice and policy. Articles examine artificial intelligence models undergoing psychotherapy style interactions, the rise of artificial intelligence scientists, and artificial intelligence designed antibodies approaching clinical trials. Commentaries argue for rethinking how artificial intelligence is built to support climate mitigation, call for global engagement on artificial intelligence governance, and discuss China’s efforts to lead on regulation. Several pieces probe artificial intelligence’s broader societal implications, including democracy, economic bubbles, research funding, neurotechnology driven mind reading, and the design of artificial intelligence centric universities. Profiles and interviews cover figures such as Yoshua Bengio and Liang Wenfeng, while Nature’s editors curate outlooks on the future of artificial intelligence and its role in mathematics, physics discovery and scientific productivity. Together, these items position computational biology and bioinformatics within a wider landscape in which artificial intelligence, data intensive methods and ethical debates are tightly coupled.
