PPI-HotSpotDB
You can download the entire PPI-HotSpotDB database as an Excel spreadsheet,
the file consists of four separate tables or tabs,
PDB_SEPARATE,
PDB_AGGREGATED,
UNIPROT and
BOUND.
- Table UNIPROT contains the most information as it contains protein chains with and without structures while PDB_SEPARATE and PDB_AGGREGATED only lists hotspots where there is an available PDB structure.
- Both PDB_SEPARATE and PDB_AGGREGATED contain the same data but PDB_AGGREGATED has all the pdb chains from the same PDB_SEPARATE file in the same row whereas PDB_SEPARATE has each pdb chain as a separate row.
- Table BOUND table lists the partner residues for any hotspot residues.
PPI-HotSpot+PDBBM
The PPI-Hotspot+PDBBM, a benchmark set of 158 (updated Sep 2023) cleaned nonredundent free PDB chains which can be downloaded below, these chains are the benchmark ones we use for testing hotspot prediction methods we have in development such as CritiRes.
- ppi_bench.tgz is a gzipped tar file containing just the single chain from each PDB in the benchmark set.
- ppi_bench.xlsx is an Excel spreadsheet version of bencmark with information such as uniprotid, pdbid, chain, residue number, the mutation, resolution, DDG and whether it is a hotspot or not.
The table below gives subset of the information on each chain in the benchmark, if you would like to use this data then we suggest obtaining it via the links above. Resolution is in Angs where "-1" indicates an NMR structure.
The PPI-HotSpotDB database and server was created by Chen Yao-Chi, Chen Yu-Hsien, Jon D Wright and Carmay Lim and is hosted at The Institute of Biomedical Sciences, Academia Sinica, Taipei 11529, Taiwan.