ATLAS: a scverse-compatible package for multi-omic single-cell trajectory inference integration
Single-cell trajectory inference is widely used to study cellular differentiation and fate decisions, yet most methods rely solely on transcriptomic data and therefore capture only part of the regulatory processes underlying cell-state transitions. Here we present ATLAS (Advanced Trajectory Learning from multi-omics At Single-cell resolution), a scverse-compatible Python package for trajectory inference from paired single-cell RNA-seq and ATAC-seq data. ATLAS integrates transcriptomic and chromatin accessibility information through Weighted Nearest Neighbor graphs, enabling both modalities to jointly inform pseudotime estimation, terminal-state identification, and fate probability inference within a unified multi-omic representation. Across synthetic and real datasets, ATLAS reconstructs coherent developmental trajectories, captures progressive fate commitment, and resolves biologically meaningful lineage structures, highlighting the value of multi-omic integration for characterizing cellular developmental dynamics. In addition, ATLAS enables joint analysis of transcription factor expression and accessibility-derived target-gene activity along pseudotime, providing insights into regulatory programs spanning transcriptomic and epigenomic layers that are not readily detectable from unimodal data. As a proof of concept, ATLAS recapitulates known hair follicle regulatory programs and reveals coherent multi-omic trajectories in which Lef1-associated regulatory patterns are linked to hair shaft differentiation. Overall, ATLAS provides an interoperable and biologically informative framework for studying cellular differentiation and regulatory dynamics in single-cell multi-omics experiments.