IBDmut estimates mutation and non-crossover gene conversion rates using identical-by-descent (IBD) segments shared by distant relatives within a population.

The program implements two methods called tMRCA regression and MaAF-threshold regression. tMRCA regression estimates the mutation rate by regressing the number of observed sequence mismatches within IBD segments on the segments’ estimated age. The slope of this regression reflects the rate at which new mutations accumulate through time, while the intercept captures the genotyping error rate.

This mutation rate estimate, however, may be biased due to non-crossover gene conversion events introducing high frequency standing variation in IBD regions. MaAF-threshold regression is used to simultaneously eliminate this bias and estimate the rate of non-crossover gene conversion events, exploiting the relationship between an allele’s frequency and the probability that it is involved in a gene conversion event.


Leveraging distant relatedness to quantify human mutation and gene conversion rates. P. Palamara, L. Francioli, P. Wilton, G. Genovese, A. Gusev, H. Finucane, S. Sankararaman, The Genome of the Netherlands Consortium, S. Sunyaev, P. de Bakker, J. Wakeley, I. Pe’er, A. Price. The American Journal of Human Genetics. November 2015. [paper] [preprint]