# Efficient methods for identifying cryptic relatedness in millions of individuals

> **NIH NIH R21** · DANA-FARBER CANCER INST · 2020 · $206,659

## Abstract

PROJECT SUMMARY
Utilizing large-scale bio bank studies to understand disease and health outcomes requires understanding the
fine-scale genetic relationships between individuals. Recent, fine-scale genetic relationships can be detected
using short segments that are inherited identical by descent (IBD) from a common ancestor between
purportedly “unrelated” pairs of individuals in a data set. Such IBD segments are a hallmark of cryptic
relatedness, which is expected to be ubiquitous in any large-scale human cohort and confounds genotype-
phenotype studies by inducing subtle population stratification that lead to false positive associations. At the
same time, IBD segments resulting from these relationships capture signal from rare variants and haplotypes
that are not directly assayed on genotyping arrays. Understanding IBD variation is thus critical for genome-
wide association studies, analyses of heritability, and genetic risk prediction. Here, we propose novel
computational methods to efficiently identify pairwise IBD segments for millions of individuals and accurately
quantify their detailed coalescent distributions.

## Key facts

- **NIH application ID:** 9989166
- **Project number:** 5R21HG010748-02
- **Recipient organization:** DANA-FARBER CANCER INST
- **Principal Investigator:** ALEXANDER GUSEV
- **Activity code:** R21 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2020
- **Award amount:** $206,659
- **Award type:** 5
- **Project period:** 2019-08-05 → 2022-07-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/9989166

## Citation

> US National Institutes of Health, RePORTER application 9989166, Efficient methods for identifying cryptic relatedness in millions of individuals (5R21HG010748-02). Retrieved via AI Analytics 2026-05-22 from https://api.ai-analytics.org/grant/nih/9989166. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*