Single-Cell Atlases, CNV Detection, and Metagenomics Advances

A new public atlas release improves harmonization across tissues and batches, making downstream comparative cell-state analysis more reliable. This resource enables researchers to integrate single-cell data from multiple studies and laboratories, reducing technical variation while preserving biological signals. The harmonized atlas supports discovery of cell types and states across the human body.

• Single-cell atlas paper

A new framework called PScnv provides personalized self-normalizing copy number variation detection with a hierarchical multi-phase approach. This method improves detection accuracy by learning patient-specific background signals, enabling more sensitive identification of somatic CNVs in cancer genomes.

• PScnv paper

A novel tool enables overlapping raw nanopore signals using a hash-based seeding mechanism, improving long-read assembly accuracy. This advancement helps resolve complex genomic regions and enables better haplotype phasing for structural variant detection.

• Rawsamble paper

This workflow reduces host-read contamination and speeds taxonomic profiling, helping clinical teams move faster from raw reads to actionable insights. The improved pipeline enables more rapid diagnosis of infections and better characterization of microbiome-associated conditions.

• Clinical metagenomics paper

A curated isoform resource with API access enables easier exploration of cancer-specific transcript usage for biomarker discovery and validation. This resource provides comprehensive isoform annotations specifically for cancer contexts, supporting identification of biomarker candidates and therapeutic targets.

• Cancer isoform resource