Haplogroups

Mitochondria Haplogroups

There are several different programs available to find out the mitochondrial haplogroup of a sample (see ...), but the two most commonly used by us are: Haplofind and HaploGrep.

mtconsensus_angsd.sh <BAMFILE> <OUTNAME>

java -jar haplogrep-2.1.16.jar --in /proj/snic2020-2-10/nobackup/private/Data/Human/Ancient/mt_consensus_strict/${file} --format fasta --out ${file}.haplogroups.txt

Y Chromosome Haplogroups

For the definition of haplogroups, Torsten has prepared 3 different lists of biallelic SNP sites from PhylotreeY (20160309):

/proj/snic2020-2-10/private/Data/Human/Ancient/bin/Ytyping/Y.snps.Phylo.noIndel.bed - all single base subsitutions from Phylotree

/proj/snic2020-2-10/private/Data/Human/Ancient/bin/Ytyping/Y.snps.Phylo.noPMD.noIndel.bed - the same as above but without transitions since those could be post mortem damage

/proj/snic2020-2-10/private/Data/Human/Ancient/bin/Ytyping/Y.snps.Phylo.noPMD.noIndel.noStrand.bed - the same as above but also A/T and G/C sites are removed to exclude strand misidentification

The first file is obviously the most comprehensive regarding haplogroup information. However, a single damaged site may skrew up things and lead to conflicts in the classification. An alternative to using the “noPMD” version would be to rescale the base qualities of Ts and As close to fragment ends (see MapDamage2.0). That is not part of our normal sequence processing but you might want to do it for your specific sample. Furthermore, I have noticed that some sites in the Phylotree list may be defined for the wrong strand (example: I found one marker CTS616 which defines I2a2a1 derived in samples where no other SNP indicated an I, so I suspect that something went wrong since it’s a C->G SNP), so such sites should be treated with caution - hence the “noStrand” file.

Here is how to use it:

1. Log in to Uppmax and start an interactive session.

2. Load the modules bioinfo-tools and samtools/1.3

3. Create a pileup file of your sample’s Y chromosome at the sites of interest:

   samtools mpileup -B -q 30 -Q 30 -f /proj/snic2020-2-10/private/Data/Human/Ancient/ref_seqs/hs37d5.fa -l <insert the BED file of your choice> <insert BAM file of your choice> > sample.out

   Note

   1. This only works for BAM files mapped to the hg19 reference genome.

   2. The last part “> sample.out” belongs to the command to write the output into a file called sample.out while the two “<>” statements need to be replaced by file names and their absolute paths (you can replace sample.out with any other file name, but you’ll need the ‘>’).

   3. -q and -Q denote minimum mapping and base quality, 30 is our usual cutoff but it might be a little stringent.

4. Run the script Ytyper.py to merge the pileup with the Phylotree haplogroup definitions:

   Ytyper.py <sample.out> (sample.out obviously needs to be replaced by your mpileup outfile’s name)

   This will produce a file sample.out.merge.txt which has one line per mutation that got called in your individual. The file is tab-separated and each column has a title. Please note that the reference sequence does not always match the ancestral allele. Commas and dots in the mpileup column indicate reads identical to the reference, alternative bases are shown as letters (see more details on the mpileup codes here). Each line is annotated as ancestral, derived or conflict (either two different alleles at that site (=contamination or sequencing/mapping error) or an allele that differs from the two alleles in Phylotree (=sequencing/mapping error or damage)).

Bibliography