Biology subjects

Gunsalus, L.

Publications and source records attributed to Gunsalus, L..

4 recordsLinked to original sources

bioRxiv2026

Agentic systems are adept at solving well-scoped, verifiable problems in computational biology

We introduce CompBioBench, a benchmark of 100 diverse tasks for evaluating agentic systems in computational biology. Unlike mathematics and programming, which more readily admit systematic verification, biological data are inherently noisy and open to interpretation. To enable objective evaluation without reducing tasks to prescriptive checklists, we propose a new benchmark construction strategy based on synthetic/augmented data and metadata scrambling/scrubbing of real datasets to create challenging problems that have a single ground-truth answer and require multi-step reasoning, tool use, bespoke code, and interaction with real-world external resources. The benchmark spans genomics, transcriptomics, epigenomics, single-cell analysis, human genetics, and machine learning workflows. Questions are curated by domain experts to cover a broad range of skills with varying difficulty. We evaluate leading general-purpose agentic systems starting from a bare-minimum environment, requiring them to fetch data and tools as needed to solve each problem. We find strong end-to-end performance, with Codex CLI (GPT 5.4) reaching 83% accuracy, Gemini CLI (3.1 Pro) reaching 82%, Claude Code (Opus 4.6) reaching 81%, and Claude Code (Opus 4.7) reaching 78%. On the hardest questions, Claude Code (Opus 4.6) reaches 69%, Codex CLI (GPT 5.4) reaches 59%, and Gemini CLI (3.1 Pro) reaches 49%. CompBioBench provides a practical testbed for measuring the progress of agentic systems in computational biology and for guiding future benchmark design. Data and a public leaderboard are available at https://huggingface.co/collections/Genentech/compbiobench-v1.

bioinformatics↗

bioRxiv2024

Decoding sequence determinants of gene expression in diverse cellular and disease states

Sequence-to-function models that predict gene expression from genomic DNA sequence have proven valuable for many biological tasks, including understanding cis-regulatory syntax and interpreting non-coding genetic variants. However, current state-of-the-art models have been trained largely on bulk expression profiles from healthy tissues or cell lines, and have not learned the properties of precise cell types and states that are captured in large-scale single-cell transcriptomic datasets. Thus, they lack the ability to perform these tasks at the resolution of specific cell types or states across diverse tissue and disease contexts. To address this gap, we present Decima, a model that predicts the cell type- and condition- specific expression of a gene from its surrounding DNA sequence. Decima is trained on single-cell or single-nucleus RNA sequencing data from over 22 million cells, and successfully predicts the cell type-specific expression of unseen genes based on their sequence alone. Here, we demonstrate Decimas ability to reveal the cis-regulatory mechanisms driving cell type-specific gene expression and its changes in disease, to predict non-coding variant effects at cell type resolution, and to design regulatory DNA elements with precisely tuned, context-specific functions.

genomics↗

bioRxiv2024

gReLU: A comprehensive framework for DNA sequence modeling and design

Deep learning models are increasingly being used to perform a variety of tasks on DNA sequences, such as predicting tissue- and cell type-specific sequence activity, deriving cis-regulatory rules, predicting non-coding variant effects, and designing synthetic regulatory sequences. However, these models require specialized knowledge to build, train and interpret correctly. In addition, the field is hampered by the lack of interoperability between models and software built by different groups. Here we present gReLU, a comprehensive software framework that enables users to easily execute advanced sequence modeling pipelines, including data preprocessing, model training, hyperparameter tuning, evaluation, interpretation, variant effect prediction, and design of novel regulatory elements. The software is accompanied by a model zoo containing state-of-the-art pre-trained models that can easily be downloaded, applied, and fine-tuned. This framework and resources will accelerate research across the field of DNA sequence modeling and enable the effective design of synthetic regulatory elements.

genomics↗

bioRxiv2020

Cell type-specific histone acetylation profiling of Alzheimer's Disease subjects and integration with genetics

We profile genome-wide histone 3 lysine 27 acetylation (H3K27ac) of 3 major brain cell types from hippocampus and dorsolateral prefrontal cortex (dlPFC) of subjects with and without Alzheimers Disease (AD). We confirm that single nucleotide polymorphisms (SNPs) associated with late onset AD (LOAD) prefer to reside in the microglial histone acetylome, which varies most strongly with age. We observe acetylation differences associated with AD pathology at 3,598 peaks, predominantly in an oligodendrocyte-enriched population. Strikingly, these differences occur at the promoters of known early onset AD (EOAD) risk genes (APP, PSEN1, PSEN2, BACE1), late onset AD (LOAD) risk genes (BIN1, PICALM, CLU, ADAM10, ADAMTS4, SORL1 and FERMT2), and putative enhancers annotated to other genes associated with AD pathology (MAPT). More broadly, acetylation differences in the oligodendrocyte-enriched population occur near genes in pathways for central nervous system myelination and oxidative phosphorylation. In most cases, these promoter acetylation differences are associated with differences in transcription in oligodendrocytes. Overall, we reveal deregulation of known and novel pathways in AD and highlight genomic regions as therapeutic targets in oligodendrocytes of hippocampus and dlPFC.

genetics↗