You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
My guess is that this is a different encoding scheme. But this is different from the spec at ftp://ftp.ncbi.nlm.nih.gov/pubchem/specifications/pubchem_fingerprints.txt, which makes it hard to connect this fingerprint to the actual properties.
The text was updated successfully, but these errors were encountered:
the function above outputs 900 bits, but Pubchem specifications say that, fingerprints are 881 bits in length ( http://astro.temple.edu/~tua87106/list_fingerprints.pdf). Therefore, how can I get 881 bits format from pubchempy’s fingerprint function?
The specifications for the compound fingerprints suggest different coding then the program reports.
For example:
CID 7844's fingerprint should be:
'AAADccBgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAGAAAAAAACACAAAACAAAAAACAACBCAAAAAAAgAAAIAAAAAAgAAAAAAQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA'
This is according to both the PubChem spec and the PubChem download page.
However the fingerprint collected from the program is:
'00000371C06000000000000000000000000000000000000000000000000000000000000000000018000000000008008000000200000000008000204200000000002000000800000000080000000001000000000000000000000000000000000000000000000000000000000000000000000000'
My guess is that this is a different encoding scheme. But this is different from the spec at ftp://ftp.ncbi.nlm.nih.gov/pubchem/specifications/pubchem_fingerprints.txt, which makes it hard to connect this fingerprint to the actual properties.
The text was updated successfully, but these errors were encountered: