README.md

Example data set

Description

This is an example data set of N=2000 individuals genotyped at G=4000 SNPs. The true number of ancestries K is equal to 3. The accuracy of the estimates should be lower because the number of individuals is small.

Prerequisites

1. Build MESA from source at the project root.

Situations

Consider phenotypes

1. Suppose the number of threads you want to allocate for MESA is 4. Run

   export OMP_NUM_THREADS=4
   ./build/mesa \
       -i ./example/N2000_K3_G4000_twisted_f6bfef.bed \
       -N 2000 \
       -K 3 \
       -G 4000 \
       --cov ./example/N2000_K3_G4000_twisted_f6bfef_R2_cov.tsv \
       -o ./example/N2000_K3_G4000_twisted_f6bfef
   

   to fit a K=3 model to the example data set.

2. If MESA runs successfully, it will produce four files

   ./example/N2000_K3_G4000_twisted_f6bfef_p.tsv
   ./example/N2000_K3_G4000_twisted_f6bfef_q.tsv
   ./example/N2000_K3_G4000_twisted_f6bfef_effect.tsv
   ./example/N2000_K3_G4000_twisted_f6bfef_effect_se.tsv
   

Do not consider phenotypes

1. Suppose the number of threads you want to allocate for MESA is 4. Run

   export OMP_NUM_THREADS=4
   ./build/mesa \
       -i ./example/N2000_K3_G4000_twisted_f6bfef.bed \
       -N 2000 \
       -K 3 \
       -G 4000 \
       -o ./example/N2000_K3_G4000_twisted_f6bfef
   

   to fit a K=3 model to the example data set.

2. If MESA runs successfully, it will produce two files

   ./example/N2000_K3_G4000_twisted_f6bfef_p.tsv
   ./example/N2000_K3_G4000_twisted_f6bfef_q.tsv