AllenInstitute/DNA-sequence-model

1---2license: mit3tags:4- neuroscience5- genomics6- dna7- sequence-modeling8- basal-ganglia9- spinal-cord10- astrocyte11- regulatory-genomics12pretty_name: DNA Sequence Modeling for BG Cell Atlas Package13---14 15# DNA Sequence Modeling for the Basal Ganglia Cell Atlas Package16 17This repository contains DNA-sequence modeling resources associated with the basal ganglia (BG) cell atlas package. It serves as a centralized entry point for sequence-based regulatory analyses across multiple companion studies.18 19## What this repository is20 21This repository is a landing page and resource hub for DNA-sequence modeling analyses associated with the basal ganglia cell atlas package.22 23## What this repository is not24 25This repository is not intended to duplicate the full analysis pipelines of the companion studies. Instead, it provides a centralized entry point for sequence-modeling-related resources, outputs, and links.26 27## Overview28 29Across these studies, sequence-based modeling is used to investigate cis-regulatory logic underlying cell type specialization. These analyses complement multiomic profiling (snRNA-seq, snATAC-seq, spatial transcriptomics, methylation, and chromatin conformation) by providing a sequence-level perspective on candidate enhancers and regulatory programs.30 31## Associated studies32 33### 1. Basal ganglia consensus taxonomy  34**Johansen, Fu et al., 2025**35 36This study establishes a consensus basal ganglia taxonomy by integrating HMBA single-nucleus RNA-seq data from human, macaque, marmoset, and previously published mouse datasets. The resulting framework provides a standardized naming system for basal ganglia cell types, enabling cross-species comparison, community-wide adoption, and downstream tool development.37 38### 2. Cross-species spinal cord atlas  39**Schmitz, Johansen et al., 2026**40 41This study presents a unified cross-species atlas integrating single-nucleus multiomic profiling and spatial transcriptomics across human, macaque, and mouse. In addition to defining a conserved cell type hierarchy, it links molecular identities to anatomical organization and cis-regulatory programs, including sequence-based modeling of enhancer logic.42 43### 3. Human BG astrocyte subgroup study  44**Fu et al., 2026**45 46This study identifies three major astrocyte subgroups in the human basal ganglia and characterizes their spatial, molecular, and regulatory specialization. Sequence-based modeling is used to evaluate subgroup-associated regulatory elements and candidate enhancer programs.47 48## Repository contents49 50Depending on the final organization, this repository include:51- model configuration files  52- input sequence sets (candidate regulatory regions)  53- cell type-enriched regions  54- model predictions and scoring outputs  55- motif and enhancer-level summaries  56- example loci used in the manuscript  57 58## Conceptual workflow59 601. Define candidate regulatory regions from multiomic data  612. Extract DNA sequences for modeling  623. Train or apply sequence-based models  634. Score sequences for regulatory activity  645. Interpret subgroup- or cell type-specific regulatory patterns  65 66## Links67 68- Basal ganglia consensus taxonomy paper: [add link]  69- Spinal cord consensus atlas paper: [add link]  70- Basal ganglia astrocyte study: [add link]  71- Multiomic track viewer (SCMDAP): [add link]  72- Project GitHub repository: [add link]  73 74## Citation75 76Please cite the relevant companion manuscripts when using these resources.77 78@article{BG_PACKAGE,79  title   = {Cross-species consensus atlas of the primate basal ganglia},80  author  = {Johansen, Nelson and Fu, Yuanyuan and others},81  journal = {bioRxiv},82  year    = {2025}83}84 85@article{BG_PACKAGE,86  title   = {A consensus spinal cord cell type atlas across mouse, macaque, and human},87  author  = {Schmitz, Matthew and Johansen, Nelson and others},88  journal = {bioRxiv},89  year    = {2026}90}91 92@article{BG_PACKAGE,93  title   = {Circuit-dependent specialization of human basal ganglia astrocytes},94  author  = {Fu, Yuanyuan and others},95  journal = {bioRxiv},96  year    = {2025}97}98