Calculate HRD Score (beta)
The Calculate HRD Score (beta) tool is designed to calculate Homologous Recombination Deficiency (HRD) from targeted research resequencing experiments.
The tool takes a target-level ploidy track (from the Detect Regional Ploidy tool) and centromers as input.
Calculate HRD Score (beta) is available from the Tools menu at:
Tools | Biomedical Genomics Analysis (
) | Oncology Score Estimation (
) | Calculate HRD Score (beta) (
)
Select the target-level ploidy track generated by the Detect Regional Ploidy tool and click Next.
You are now presented with choices regarding HRD calculation.
- Centromeres Centromeres are used to define the chromosome arms and must be provided for correct calculation of the HRD score, as LOH, LST and TAI events are calculated per chromosome arm. Regions covered by provided centromeres are excluded from the HRD calculation.
- LOH weight, LST weight, TAI weight HRD is calculated as the sum of Loss of Heterozygosity (LOH), Large-scale State Transitions (LST), and Telomeric Allelic Imbalance (TAI) scores. These three scores can be weighted by the values given here before being summed to the final HRD score.
- Minimum LOH region length (MB) Loss of Heterozygosity (LOH) regions shorter than this are ignored in the LOH score calculation. The length is given in megabases.
- Minimum LST region length (MB) Large-scale State Transitions (LST) between regions shorter than this are ignored in the LST score calculation. The length is given in megabases.
- Maximum LST region distance (MB) Large-scale State Transitions (LST) with a distance larger than this are ignored in the LST score calculation. The distance is given in megabases.
- Short LST region length (MB) Regions shorter than this are filtered out before Large-scale State Transitions (LST) score calculation. The length is given in megabases.
- Minimum merge size Remove merged regions consisting of fewer than this number of targets.
When finished with the settings, click Next to start the algorithm.
HRD calculation
The HRD score is a count of chromosomal rearrangements that can be increased in tumors with HRD. It is calculated as the weighted sum of three different chromosomal rearrangements: The number of Telomeric Allelic Imbalances (TAI), Large-scale Transitions (LST), and long regions of Loss of Heterozygosity (LOH). The calculations are based on identified regions of copy number variations (CNV) as well as variant frequencies in a sample, which are identified beforehand.
The LOH score is defined as the number of regions with a minor allele count of zero. If all regions on a chromosome are affected by LOH, the whole chromosome is excluded from the LOH score calculation.
The TAI score is defined as the number of regions that:
- (1) Show an imbalance from the most prevalent copy number state of the whole chromosome. Whether a telomeric region is imbalance with the rest of the chromosome, is determined by comparing the copy number state of the region nearest the telomere with the most prevalent copy number state of the whole chromosome.
- (2) Do not cross the centromere.
- (3) Extend up to the telomeres. Note, the region located closest to the end of a chromosome is considered a proxy for the telomeric region. The CNV regions underlying TAI are filtered for a minimum number of probes (merge size) and a minimum region length.
Calculation of the three scores is inspired by:
- Abkevich et al. Patterns of genomic loss of heterozygosity predict homologous recombination repair defects in epithelial ovarian cancer, British Journal of Cancer. 2012, 107(10): 1776-1782. [Abkevich et al., 2012]
- Birkbak et al. Telomeric allelic imbalance indicates defective DNA repair and sensitivity to DNA damaging agents, Cancer Discovery. 2012, 2(4): 366-375. [Birkbak et al., 2012]
- de Luca et al. Using whole-genome sequencing data to derive the homologous recombination deficiency scores. npj Breast Cancer. 2020, 6:33. [de Luca et al., 2020]
The LST score is the number of LST events. The LST score counts large rearrangements for each arm of a chromosome. The regions are merged and short regions removed iteratively. For each chromosome arm, as long as there are segments less than 3 MB, the segment at the first position, that is less than 3 MB is removed and adjacent segments across the whole chromosome arm with identical allele counts merged.
Calculate HRD score algorithm report
The report provides a table listing:
- HRD score The HRD score calculated from the LOH, LST and TAI scores.
- HRD score LOH/LST/TAI weight The weight that each of the underlying scores were given when calculating the HRD score.
- Genomic coverage The percentage of genomic positions covered by the panel used for HRD detection.
- LOH score The number of LOH regions.
- LOH regions The positions of the identified LOH regions. For each region, the chromosome and the start and end of the LOH region is included. As an example, the entry "2: 151M 169M" should be read as an LOH event on chromosome 2 occurring from position 51M to 169M.
- LST score The number of LST regions.
- LST The positions of the identified LST regions. As an example, in the entry "S1: 1-2 0M 13M -> 1-1 13M 248M" the parts before and after the arrow describe the chromosomal states on each side of the transition and should be read as: Start of chromosome 1, minor allele count 1, major allele count 2, positions 0M-13M changes to minor allele count 1, major allele count 1, positions 13M to 248M.
- TAI score The number of TAI regions.
- TAI The positions of the identified TAI regions. As an example "S1 TAI 2 1-2", should be read as start of chromosome 1, TAI event, most prevalent copy number state for the whole chromosome is 2, for the TAI event minor allele count is 1 and major allele count is 2. Correspondingly, "E1 CENT 125M 248M" should be read as end of chromosome 1, region extends from end of chromosome to centromere and is not counted as TAI, positions 125M-248M and "E10 NO 2 1-1" should be read as end of chromosome 10, no TAI event, most prevalent copy number state for the whole chromosome is 2, and for the region closest to the end of the chromosome minor allele count is 1 and major allele count is 1. Hence, a TAI event is only counted when TAI is part of the annotation for a given chromosome arm.