Description

This track collection shows Combined Annotation Dependent Depletion scores. CADD is a tool for scoring the deleteriousness of single nucleotide variants as well as insertion/deletion variants in the human genome.

Some mutation annotations tend to exploit a single information type (e.g., phastCons or phyloP for conservation) and/or are restricted in scope (e.g., to missense changes). Thus, a broadly applicable metric that objectively weights and integrates diverse information is needed. Combined Annotation Dependent Depletion (CADD) is a framework that integrates multiple annotations into one metric by contrasting variants that survived natural selection with simulated mutations.

CADD scores strongly correlate with allelic diversity, pathogenicity of both coding and non-coding variants, experimentally measured regulatory effects, and also rank causal variants within individual genome sequences with a higher value than non-causal variants. Finally, CADD scores of complex trait-associated variants from genome-wide association studies (GWAS) are significantly higher than matched controls and correlate with study sample size, likely reflecting the increased accuracy of larger GWAS.

A CADD score represents a ranking not a prediction, and no threshold is defined for a specific purpose. Higher scores are more likely to be deleterious: Scores are

  10 * -log of the rank

Display Conventions and Configuration

There are six subtracks of this track: four for single-nucleotide mutations, one for each base, showing all possible substitutions, one for insertions and one for deletions. All subtracks show the CADD Phred score on mouseover. Zooming in shows the exact score on mouseover, same basepair = score 0.0.

PHRED-scaled scores are normalized to all potential ~9 billion SNVs, and thereby provide an externally comparable unit for analysis. For example, a scaled score of 10 or greater indicates a raw score in the top 10% of all possible reference genome SNVs, and a score of 20 or greater indicates a raw score in the top 1%, regardless of the details of the annotation set, model parameters, etc.

The four single-nucleotide mutation tracks have a default viewing range of score 10 to 50. As explained in the paragraph above, that results in slightly less than 10% of the data displayed. The deletion and insertion tracks have a default filter of 10-100, because they display discrete items and not graphical data.

Single nucleotide variants (SNV): For SNVs, at every genome position, there are three values per position, one for every possible nucleotide mutation. The fourth value, "no mutation", representing the reference allele, e.g., A to A, is always set to zero.

When using this track, zoom in until you can see every basepair at the top of the display. Otherwise, there are several nucleotides per pixel under your mouse cursor and instead of an actual score, the tooltip text will show the average score of all nucleotides under the cursor. This is indicated by the prefix "~" in the mouseover. Averages of scores are not useful for any application of CADD.

Insertions and deletions: Scores are also shown on mouseover for a set of insertions and deletions. On hg38, the set has been obtained from gnomAD3. On hg19, the set of indels has been obtained from various sources (gnomAD2, ExAC, 1000 Genomes, ESP). If your insertion or deleletion of interest is not in the track, you will need to use CADD's online scoring tool to obtain them.

Data access

CADD scores are freely available for all non-commercial applications from the CADD website. For commercial applications, see the license instructions there.

The CADD data on the UCSC Genome Browser can be explored interactively with the Table Browser or the Data Integrator. For automated download and analysis, the genome annotation is stored at UCSC in bigWig and bigBed files that can be downloaded from our download server. The files for this track are called a.bw, c.bw, g.bw, t.bw, ins.bb and del.bb. Individual regions or the whole genome annotation can be obtained using our tools bigWigToWig or bigBedToBed which can be compiled from the source code or downloaded as a precompiled binary for your system. Instructions for downloading source code and binaries can be found here. The tools can also be used to obtain features confined to a given range, e.g.,
bigWigToBedGraph -chrom=chr1 -start=100000 -end=100500 http://hgdownload.soe.ucsc.edu/gbdb/hg38/cadd/a.bw stdout
or
bigBedToBed -chrom=chr1 -start=100000 -end=100500 http://hgdownload.soe.ucsc.edu/gbdb/hg38/cadd/ins.bb stdout

Methods

Data were converted from the files provided on the CADD Downloads website, provided by the Kircher lab, using custom Python scripts, documented in our makeDoc files.

Credits

Thanks to the CADD development team for providing precomputed data as simple tab-separated files.

References

Kircher M, Witten DM, Jain P, O'Roak BJ, Cooper GM, Shendure J. A general framework for estimating the relative pathogenicity of human genetic variants. Nat Genet. 2014 Mar;46(3):310-5. PMID: 24487276; PMC: PMC3992975

Rentzsch P, Witten D, Cooper GM, Shendure J, Kircher M. CADD: predicting the deleteriousness of variants throughout the human genome. Nucleic Acids Res. 2019 Jan 8;47(D1):D886-D894. PMID: 30371827; PMC: PMC6323892