DOWNSAMPLING EFFECT IN HOMOLOGOUS RECOMBINATION DEFICIENCY ANALYSIS TOOLS

Rafael Lucas Muniz Guedes; Rodrigo de Souza Reis; Paulo Campregher; João Bosco Oliveira Filho; Murilo Cervato

All Papers

Submission

Paper Title

DOWNSAMPLING EFFECT IN HOMOLOGOUS RECOMBINATION DEFICIENCY ANALYSIS TOOLS

Authors

Rafael Lucas Muniz Guedes
Rodrigo de Souza Reis
Paulo Campregher
João Bosco Oliveira Filho
Murilo Cervato

Modality

Xpress presentation

Subject area

Omics

Publishing Date

26/04/2022

Country of Publishing

Brasil

Language of Publishing

Inglês

Paper Page

https://www.even3.com.br/anais/xmeetingxp2021/414885-downsampling-effect-in-homologous-recombination-deficiency-analysis-tools

ISBN

978-65-5941-645-5

Keywords

Homologous Recombination Deficiency, WGS, Downsampling

Summary

Genomic instability is a common trait in many types of cancer, such as breast, ovary and prostate. It is characterized by the accumulation of DNA mutations during successive cell divisions, ranging from single nucleotide polymorphisms to gross chromosomal changes. The presence of dysfunctional genes in the homologous recombination pathway, like BRCA1/2, is considered one of the main causes of instability known as Homologous Recombination Deficiency (HRD). Due to its growing importance as a biomarker and availability of FDA approved treatments, different metrics have been proposed in order to detect HRD from whole genome sequencing (WGS) samples. The cost of WGS sequencing can be a limiting factor for a faster and broader application of HRD genetic tests in clinical practice, therefore, in the present study, we conducted a preliminary analysis of the effect of downsampling sequencing coverage in the outcome of five tools for HRD analysis, which are mainly based in the detection of loss of heterozygosity, large-scale state transition, telomeric allelic imbalance or mutational signatures. Six WGS paired breast cancer samples (GCRC1735, GCRC1738, GCRC1828, GCRC1887, GCRC145 and GCRC2001) were evaluated with five tools with 40X (original) and 10X (downsample) mean coverage. The following thresholds were considered to categorize samples as high (HI) or low instability (LI) for each metric: HRDetect (HI: HRDetect score >= 0.7), scarHRD (HI: scar sum >= 42), CHORD (HI: p_hrd >= 0.5), decompTumor2Sig (HI: SBS3 signature >= 0.3) and shalowHRD (HI: LGA >= 20; LI: LGA <15; borderline: 15 <= LGA <= 19). HRDetect, scarHRD and CHORD presented, respectively 2, 1 and 2 samples with category shift after downsampling. On average, only 30.4% of original SNPs were retrieved from 10X samples, although no category change for SBS3 mutational signatures was observed for DecompTumor2Sig. Regarding shallowHRD, the final scores for three samples was unchanged and alteration of a single unit was observed for the remaining samples. In general, samples considered as LI are likely to be evaluated as false positives while HI with high scores are less prone to be affected by lower coverages. The results of the tests are used as a part of an ongoing validation process to identify breast and ovarian cancer patients with positive homologous recombination deficiency (HRD) status, who are eligible for treatment with the targeted therapy. ShallowHRD was the most consistent tool and is a good candidate for low-cost validation in clinical tests, with the advantage of no necessity of paired normal sample sequencing. Tests with more samples from different types of cancer are still required and will be conducted in the future.

Title of the Event: X-Meeting XPerience 2021
Title of the Proceedings of the event: X-Meeting presentations
Name of the Publisher: Even3
Means of Dissemination: Meio Digital

How to cite

GUEDES, Rafael Lucas Muniz et al.. DOWNSAMPLING EFFECT IN HOMOLOGOUS RECOMBINATION DEFICIENCY ANALYSIS TOOLS.. In: X-Meeting presentations. Anais...São Paulo(SP) AB3C, 2021. Available in: https//www.even3.com.br/anais/xmeetingxp2021/414885-DOWNSAMPLING-EFFECT-IN-HOMOLOGOUS-RECOMBINATION-DEFICIENCY-ANALYSIS-TOOLS. Access in: 06/10/2025