- See the Setup file, this:
- We produced a human reference SH2 CSV file. We hand annotated/identified differences (as noted in the notebook)
- We produced a human reference SH2 Fasta file
- We used promals3d to align
- We produced a structure reference file for all PDB IDs for the uniprot IDs from our reference, then annotated that structure reference file with UniprotRefFile in order to get inclusive domains.
- We filtered to relevant PDBs that spanned SH2 domains within them.
- AlphaFold
- CIF file fetch
- Adjacency