Btz

I wrote a tool for looking up molecules on the BTZ (Bigger than ZINC) db. The idea is the have an api that can tell you which database in SmallWorld (i.e s enamine, m enamine, wuxi) the molecule belongs to. At the moment, it only support molecule id search. Smiles search is on the work. Please note that I'm still loading the molecules into BTZ, so many molecules might not be up there yet. Please let me know your feedbacks and issues.

Example (it works with Z, PV, s_ codes from enamines): curl http://gimel.compbio.ucsf.edu:5022/api/_search_btz?molecule_id=s_270004__7454808__7549222

Output

{

  "db_name": "S-Enamine-20Q1-797M", 
  "mol_id": "s_270004__7454808__7549222", 
  "query_type": "mol_id", 
  "smiles": "Cc1cc2cc(CN3CCN(C)CC3c3ccccc3)oc2cc1C", 
  "table_name": "s_enamine"
}

Search in bulk

To do search from a list of molecule in text file and output in json and text format

## Search molecules with txt output
curl -X POST -F "input_file=@/nfs/exa/work/khtang/btz_api/mol_id.txt" http://gimel.compbio.ucsf.edu:5022/api/_bulk_search_btz?output='txt'

## Search molecules with json output
curl -X POST -F "input_file=@/nfs/exa/work/khtang/btz_api/mol_id.txt" http://gimel.compbio.ucsf.edu:5022/api/_bulk_search_btz?output='json'

As of April 13, 2020:

There are 10,107,024,050 codes in m_enamine. There will eventually be about 11 billion.

There are 1,542,514,329 codes in s_enamine. There will eventually be about 11 billion

There are 126,471,523 codes in mcule_ultimate library

We are working on how to load it faster.