Description

The Consortium of Long Read Sequencing (CoLoRSdb) catalogs both small genetic variants (single nucleotide polymorphisms and short insertions and deletions) and structural variants (insertions, deletions, and inversions) discovered from long-read whole genome sequencing. Long reads improve sensitivity and breakpoint resolution compared to short-read data, especially for repeats and complex loci.

For more information, see the CoLoRSdb website: https://colorsdb.org/

Display Conventions and Configuration

Small variants (DeepVariant + GLnexus)

• DeepVariant + GLnexus tracks are shown using the UCSC Genome Browser VCF display format.
• In dense mode, each variant appears as a single pixel.
• In pack or full modes, clicking a variant reveals the reference and alternate alleles (REF/ALT), the quality score (QUAL), the filter status (FILTER), and fields from the INFO column.
• More details on the VCF display format are available at: https://genome.ucsc.edu/goldenPath/help/vcf.html

Structural variants (pbsv + Jasmine)

• Structural variants are rendered as intervals with spans and block structures when applicable.
• Coloring for each item is based on the variant type:
  • Deletion (DEL) = red
  • Insertion (INS) = blue
  • Inversion (INV) = purple
• Mouseover an item reveals the genomic position, structural variant type, variant length, allele frequency, alternative allele count, number of samples, total allele number, Hardy-Weinburg equilibrium (HWE) p-value, and excess heterozygosity (ExcHet) p-value.

Methods

• Small variants: Per-sample variant calling was performed with DeepVariant, followed by joint genotyping across the cohort using GLnexus.
• Structural variants: Called from long-read sequencing data using pbsv and merged using Jasmine across individuals. The VCF file was converted to a bigBed file for efficient display.

Data Access

The CoLoRSdb track can be explored interactively using the REST API, the Table Browser or the Data Integrator. The VCF and bigBed files files are available from our downloads directory. The original VCF files are part of the v1.2.0 dataset, and documentation is available on Zenodo: https://zenodo.org/records/14814308

Credits

Thanks to Mike Schatz, Evan Eichler, and all CoLoRSdb investigators for generating and making the data publicly available.

References

Poplin R, Chang PC, Alexander D, Schwartz S, Colthurst T, Ku A, Newburger D, Dijamco J, Nguyen N, Afshar PT et al. A universal SNP and small-indel variant caller using deep neural networks. Nat Biotechnol. 2018 Nov;36(10):983-987. DOI: 10.1038/nbt.4235; PMID: 30247488

Yun T, Li H, Chang PC, Lin MF, Carroll A, McLean CY. Accurate, scalable cohort variant calls using DeepVariant and GLnexus. Bioinformatics. 2021 Apr 5;36(24):5582-5589. DOI: 10.1093/bioinformatics/btaa1081; PMID: 33399819; PMC: PMC8023681

Kirsche M, Prabhu G, Sherman R, Ni B, Battle A, Aganezov S, Schatz MC. Jasmine and Iris: population-scale structural variant comparison and analysis. Nat Methods. 2023 Mar;20(3):408-417. DOI: 10.1038/s41592-022-01753-3; PMID: 36658279; PMC: PMC10006329

Eisfeldt J, Ameur A, Lenner F, Ten Berk de Boer E, Ek M, Wincent J, Vaz R, Ottosson J, Jonson T, Ivarsson S et al. A national long-read sequencing study on chromosomal rearrangements uncovers hidden complexities. Genome Res. 2024 Nov 20;34(11):1774-1784. DOI: 10.1101/gr.279510.124; PMID: 39472022; PMC: PMC11610602

Lake, J. A., & Consortium of Long Read Sequencing (CoLoRS). Consortium of Long Read Sequencing Database (CoLoRSdb) (v1.2.0) [Data set] Zendo. 2025 Feb 5. 10.5281/zenodo.14814308