Relate practical - Using genealogies for population genetics

Aug 20, 2026

Relate practical focuses on inferring genome-wide genealogies for the Simons Genome Diversity Project dataset of 278 modern humans using Relate binaries, bash scripts, and R. It explains how sampling locations are represented across seven regions and how to prepare input data by converting phased genotypes to the haps/sample format, setting ancestral alleles to 0, and applying masks. The practical then guides running Relate and estimating effective population sizes and split times from inferred genealogies, with additional sections on evidence for positive selection.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Overview

Relate

Data requirements and file formats

The arguments

Running Relate on the Simons Genome Diversity Project dataset

Running Relate

Effective population sizes and split times

Estimating population sizes given a genealogy

Joint fitting of population size and branch lengths

Coaleate: Inferring coalescence rates for low-coverage genomes

Detecting evidence for positive selection

Relate practical 15-06-2022 Relate practical: Using genealogies for population genetics Jasmin Rees 1 , Leo Speidel 1 ,2 ,􀀃 1 Genetics Institute, University College London, London, UK 2 Francis Crick Institute, London, UK 􀀃 Contact: leo .speidel@outlook .com In this practical, we will infer genealgies for the Simons Genome Diversity Project dataset, downloaded from [https://reichdata.hms.harvard.edu/pub/datasets/sgdp/](https://reichdata.hms.harvard.edu/pub/datasets/sgdp/. This)[. This](https://reichdata.hms.harvard.edu/pub/datasets/sgdp/. This) dataset comprises wholegenome sequencing data of 278 modern humans with sampling locations shown in Fig. 1. Latitude | ●
●
● | ● | | ●
●
● | ● ● | | ● | ●●
●
●
●●
● | | ●
●●●●
● ● ●●
● ●
●
●●●
● | ● | | ●● | | ●
●
●
● ●●
● ●●
●●
● | ●
●● | ●
● | ●
● ● | | ●
●●
● | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | | | | | ●
●
●● | ●
●● | | | | | ● | ●
●●
●● | | | | | | | | | | | ●
● | ● | | | | | | | | | | | | | | | | | | | | | | | | | ●
● | | | | | | | | | ● ● | | | | | | | | | | ●
●● | | | | | | | | ●●
●● | ● | | ●●
●
●
●
● | | | ● | | | | | | | | | | | | | ● ● | | | | | | | | | | | | | | | | | ● | | ●
● | | | | | | | | | | | | | | ●●
● | | | | | ●●
●● | | | | | | | | | | | | | | | ● | | | | | | | | | | | | | | | | | | | | Longitude region ● Africa ● America ● CentralAsiaSiberia ● EastAsia ● Oceania ● SouthAsia ● WestEurasia Figure 1: Sampling locations of the 278 modern humans in the Simons Genome Diversity Project. Samples are classi􀀌ed into seven regions shown by colours. Note 1 The data was downloaded from: - Phased genotypes: [https://sharehost.hms.harvard.edu/genetics/reich_lab/sgdp/](https://sharehost.hms.harvard.edu/genetics/reich_lab/sgdp/)[ ](https://sharehost.hms.harvard.edu/genetics/reich_lab/sgdp/)[phased_data/PS2_multisample_public](phased_data/PS2_multisample_public) - Genomic mask: [https://reichdata.hms.harvard.edu/pub/datasets/sgdp/filters/](https://reichdata.hms.harvard.edu/pub/datasets/sgdp/filters/)[ ](https://reichdata.hms.harvard.edu/pub/datasets/sgdp/filters/)[all_samples/](all_samples/) - Human ancestral genome: [ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/phase1/](ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/phase1/)[ ](ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/phase1/)[analysis_results/supporting/ancestral_alignments/](analysis_results/supporting/ancestral_alignments/) - Recombination maps: [https://mathgen.stats.ox.ac.uk/impute/1000GP_Phase3.html](https://mathgen.stats.ox.ac.uk/impute/1000GP_Phase3.html) [We then used](We then used RelateFileFormats--mode ConvertFromVcf to convert to the haps/sample)[ RelateFileFormats--mode ConvertFromVcf](We then used RelateFileFormats--mode ConvertFromVcf to convert to the haps/sample)[ to convert to the haps/sample](We then used RelateFileFormats--mode ConvertFromVcf to convert to the haps/sample) 􀀌le format and the [PrepareInputFiles.sh](PrepareInputFiles.sh) script to make sure ancestral alleles are denoted by 0 and to 􀀌lter out regions according to the genomic mask. Overview 1 Relate 2 1.1 Data requirements and 􀀌le formats ............................. 3 1.2 The arguments ........................................ 3 2 Running Relate on the Simons Genome Diversity Project dataset 4 2.1 Running Relate ........................................ 4 3 E􀀋ective population sizes and split times 6 3.1 Estimating population sizes given a genealogy ....................... 6 3.2 Joint 􀀌tting of population size and branch lengths .................... 8 3.3 Colate: Inferring coalescence rates for low-coverage genomes ............... 9

FAQs

What is Relate used for in this practical?

Relate estimates joint genealogies for many thousands of modern individuals across the genome, describing relationships through most-recent common ancestors.

How are the SGDP input data prepared for Relate?

Phased genotypes are converted to the haps/sample file format, ancestral alleles are denoted by 0, and genomic regions are filtered using the provided genomic mask.

What analyses are performed after running Relate?

The practical estimates effective population sizes and split times from inferred genealogies, including methods such as joint fitting and coalescence-rate inference for low-coverage genomes.

Aug 20, 2026

Upload by วิน