Lab journal

Multiple sequence alignment

The HFE protein consists of 3 domains: alpha 1 and 2 and a Immunoglobulin C1-set domain. Therefore we found two different pfam families that both match our protein: Class I Histocompatibility antigen, domains alpha 1 and 2 (http://pfam.sanger.ac.uk/family/PF00129) and Immunoglobulin C1-set domain (http://pfam.sanger.ac.uk/family/PF07654#tabview=tab0). We checked how much of out proteins sequence is covered by both fimilies: immunoglobulin
# GS HFE_HUMAN/211-294 DR PDB; 1A6Z A; 189-272;

mhc class 1
# GS HFE_HUMAN/26-202 DR PDB; 1A6Z A; 4-180;

We downloaded the alignments in fasta format. Therefore, we first formatted the alignment to fasta format and then downloaded the .txt file. The alignment were then formatted again with a2m2aln:

/usr/share/freecontact/a2m2aln -q '^HFE_HUMAN/(\d+)' --quiet < imm.txt > imm.aln
/usr/share/freecontact/a2m2aln -q '^HFE_HUMAN/(\d+)' --quiet < mhc.txt > mhc.aln
/usr/share/freecontact/a2m2aln -q '^RASH_HUMAN/(\d+)' --quiet < ras.txt > ras.aln

Calculate and analyze correlated mutations

freecontact with standard parameters and evfold as output format this was done with the following command:

freecontact -o evfold < imm.aln > imm_contacts.out
freecontact -o evfold < mhc.aln > mhc_contacts.out
freecontact -o evfold < ras.aln > ras_contacts.out

The contact predictions were then analysed using the following pipeline

1. Remove contacts that have a distance of <= 5 residues in sequence
2. Order descending by CN score and plot score distribution
3. Select high scoring pairs
4. Load reference PDB and determine minimum atom distance between all residue pairs
5. determine pairs that have a minimum distance of below 5 A

Ras hotspots MHC I hotspots Ig C1 hotspots