New Software for the Fast Estimation of Population Recombination Rates (FastEPRR) in the Genomic Era

Abstract

Genetic recombination is a very important evolutionary mechanism that mixes parental haplotypes and produces new raw material for organismal evolution. As a result, information on recombination rates is critical for biological research. In this paper, we introduce a new extremely fast open-source software package (FastEPRR) that uses machine learning to estimate recombination rate ρ (=4Ner) from intraspecific DNA polymorphism data. When ρ>10 and the number of sampled diploid individuals is large enough (≥50), the variance of ρFastEPRR remains slightly smaller than that of ρLDhat. The new estimate ρcomb (calculated by averaging ρFastEPRR and ρLDhat) has the smallest variance of all cases. When estimating ρFastEPRR, the finite-site model was employed to analyze cases with a high rate of recurrent mutations, and an additional method is proposed to consider the effect of variable recombination rates within windows. Simulations encompassing a wide range of parameters demonstrate that different evolutionary factors, such as demography and selection, may not increase the false positive rate of recombination hotspots. Overall, accuracy of FastEPRR is similar to the well-known method, LDhat, but requires far less computation time. Genetic maps for each human population (YRI, CEU, and CHB) extracted from the 1000 Genomes OMNI data set were obtained in less than 3 d using just a single CPU core. The Pearson Pairwise correlation coefficient between the ρFastEPRR and ρLDhat maps is very high, ranging between 0.929 and 0.987 at a 5-Mb scale. Considering that sample sizes for these kinds of data are increasing dramatically with advances in next-generation sequencing technologies, FastEPRR (freely available at http://www.picb.ac.cn/evolgen/) is expected to become a widely used tool for establishing genetic maps and studying recombination hotspots in the population genomic era.

LDJump: Estimating variable recombination rates from population genetic data

“…In contrast to Gao et al. (), our approach also works well with small sample sizes down to 10 sequences.…”

“…Table S2 in Supporting Information Appendix S1 summarizes the performance of the aforementioned methods LDhat (Auton & McVean, ), LDhelmet (Chan et al., ), FastEPRR (Gao et al., ), the first published version of LDhat(v1) (McVean et al., ), and LDJump . As shown in the Supporting Information Appendix S1 (section 1.4), segment lengths of at least 400 bp are needed for a good performance of LDJump .…”

“…Futschik, Hotz, Munk, and Sieling (2014) use a similar approach to partition DNA sequences into homogeneous segments with respect to GC content. In contrast to Gao et al (2016), our approach also works well with small sample sizes down to 10 sequences. Section 2 contains a detailed description of our proposed method.…”

“…We approximated the RMSE (root mean squared error) as our quality measure, and computed the estimation errors on an equidistant grid of 1,000 positions along the sequences. Table S2 in Supporting Information Appendix S1 summarizes the performance of the aforementioned methods LDhat (Auton & McVean, 2007), LDhelmet (Chan et al, 2012), FastEPRR (Gao et al, 2016), the first published version of LDhat (v1 Indeed, the standard deviation of the RMSE is more than 50% lower with LDJump than that of LDhelmet. With FastEPRR, approximately 57%, 5%, 4%, and 2% of the computations terminated due to errors using segment lengths of 500, 1,000, 1,500, 2,000, respectively.…”

“…With the GUI of RDP (Martin et al., ) overall patterns of recombination and testing for hot and cold spots is performed with help from LDhat (McVean et al., ). Recently, alternative fast estimates of ρ that rely on regression on sliding windows have been proposed by Lin, Futschik, and Li () and Gao, Ming, Hu, and Li (). Their software implementation is called FastEPRR and is recommended by the authors for larger samples consisting of 50 sequences or more.…”