UPDATE: Zenodo migration postponed to Oct 13 from 06:00-08:00 UTC. Read the announcement.

Dataset Open Access

Relate-estimated coalescence rates, allele ages, and selection p-values for the 1000 Genomes Project

Speidel, Leo; Forest, Marie; Shi, Sinan; Myers, Simon R.


Coalescence rates, allele ages, and p-values for evidence of positive selection calculated for 2478 samples of the 1000 Genomes Project using Relate.

We estimated the joint genealogy of all 1000 GP populations and then extracted the embedded genealogy for each population.
For the genealogy of each population, we jointly estimated the population size history and branch lengths. 
Variants segregating in more than one population therefore have correlated but different allele ages in each population.

Please refer to Speidel et al. Nature Genetics (2019) for more details or email leo.speidel@outlook.com for any queries.

Coalescence rates

The zipped directory coalescence_rates.zip contains coalescence rates for 26 populations in the 1000 Genomes Project data set.

  • The .coal files show the haploid coalescence rates, please refer to the Relate documentation for the file format.
  • The popsize.RData file is an R data frame storing the diploid population sizes (0.5/coalescence rate) calculated using the .coal files. The columns of this data frame, named "pop_size", are
    • gens_ago: Time in generations at which epoch starts. (To get years from generations, we multiply by 28.)
    • population_size: Diploid population size in this epoch.
    • population: Name of population 
    • region: Name of region (AFR, AMR, EAS, EUR, SAS)

Allele ages and selection p-values

The zipped directories allele_ages_*.zip contain R data frames for each 1000GP population storing allele ages and selection p-values.
Please note that only mutations that segregate in the population and map to a unique branch in the Relate-estimated marginal trees are included. Selection p-values are only provided for mutations of DAF > 2 that pass quality filters (see Speidel et al., 2019). 

To get an age estimate for a neutral mutation, use 0.5*(lower_age + upper_age). To get years from generations, we multiply by 28.

The columns of these data frames, named "allele_ages", are

  • CHR: chromosome index
  • BP: base-pair position (GRCh37)
  • ID: id of SNP
  • lower_age: Age in generations of coalescence event at the lower end of the branch onto which the mutation maps
  • upper_age: Age in generations of coalescence event at the upper end of the branch onto which the mutation maps
  • ancestral/derived: Ancestral/derived allele
  • upstream: Upstream (5') allele
  • downstream: Downstream (3') allele
  • DAF: Derived-allele frequency
  • pvalue: log10 p-value for selection evidence
For R object files, use load() to load data frames into R.
Files (7.4 GB)
Name Size
2.8 GB Download
1.1 GB Download
1.0 GB Download
1.2 GB Download
1.2 GB Download
19.7 kB Download
  • Speidel et al., Nature Genetics 2019, A method for genome-wide genealogy estimation for thousands of samples. https://doi.org/10.1038/s41588-019-0484-x

All versions This version
Views 2,2462,246
Downloads 2,0732,073
Data volume 2.9 TB2.9 TB
Unique views 2,0492,049
Unique downloads 686686


Cite as