From Genome Analysis Wiki
Jump to navigationJump to search

ChunkChromosome is a helper utility for minimac and MaCH. It can be used to facilitate analyses of very large datasets in overlapping slices. For information on how to put the resulting chunks back together, see this page.


ChunkChromosome expects three parameters:

  • A data file (specified with the -d command line option), listing all the markers along one chromosome. The data file can optionally include phenotype and other information, which is safely ignored.
  • A desired core chunk size, in markers (specified with the -n command line option and defaulting to 5000 markers).
  • A desired overlap between chunks, also in markers (specified with the -o command line option and defaulting to 500 markers).


Suppose you plan to run 1000 Genomes Imputation using MaCH and minimac. Typically, you'd accomplish this by running the following commands:

awk '{ if ($1 == "M") print $2; }' < chr1.dat > chr1.snps
mach1 -d chr1.dat -p chr1.ped --rounds 20 --states 200 --phase --prefix chr1.haps
minimac --refHaps 1000genomes.chr1.haps.gz  --refSnps 1000genomes.chr1.snps --haps chr1.haps.gz --snps chr1.snps --rounds 5 --states 200 --prefix chr1.imputed

These commands would haplotype (with MaCH) and then impute (with Minimac) an entire chromosome. While the process works, it can be rather time consuming for large chromosomes and large numbers of individuals. ChunkChromosome allows the process to be streamlined by running different portions of each chromosome in parallel.


@ length = 2500
@ overlap = 500

# Estimate haplotypes for all individuals, in 2500 marker chunks, with 500 marker overhang
foreach chr (`seq 1 22`)

   ChunkChromosome -d chr$chr.dat -n $length -o $overlap

   foreach chunk (chunk*-chr$chr.dat)

      mach -d $chunk -p chr$chr.ped --prefix ${chunk:r} \
           --rounds 20 --states 200 --phase --sample 5 >& ${chunk:r}-mach.log &



# Impute into phased haplotypes
foreach chr (`seq 1 22`)

   foreach chunk (chunk*-chr$chr.dat)

      set haps = /data/1000g/hap/all/20101123.chr$chr.hap.gz
      set snps = /data/1000g/snps/chr$chr.snps

      minimac --refHaps $haps --refSnps $snps --rounds 5 --states 200 \
              --haps ${chunk:r}.gz --snps ${chunk}.snps  --autoClip autoChunk-chr$chr.dat  \
              --prefix ${chunk:r}.imputed >& ${chunk:r}-minimac.log &



The autoChunk file, generated by the ChunkChromosome program, tells minimac what are the markers of interest for each chunk. This allows chunks to overlap (which improves accuracy near the edges) but still ensures that each marker is only imputed once.


You can download source code for the ChunkChromosome program in a tar-ball archive generic-ChunkChromosome-2014-05-27.tar.gz. After downloading it, unpack the archive and use Make to compile the tool.