Skip to content
a molecular encoder
OCaml Python Shell Makefile
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
bin
data
doc
src
LICENSE
Makefile
README.md
TODO
dune-project
kb_test.sh
molenc.opam
tani_est.gpl
test.sh

README.md

molenc

Molecular encoder using rdkit and OCaml.

The implemented fingerprint is J-L Faulon's "Signature Molecular Descriptor". This is an unfolded-counted chemical fingerprint. Such fingerprints are less lossy than most well-known chemical fingerprints like ECFP4.

Some advantages: such fingerprints don't create feature collisions upon encoding. Also, upon encoding, a feature dictionary is created. It can be used later on to map a given feature index to an atom environment.

We recommend using a radius of zero to one (molenc_d -r 0:1 ...).

The fingerprint can be run using atom types (#pi-electrons, element symbol, #HA neighbors, formal charge) or rdkit pharmacophore features (TODO) (Donor, Acceptor, PosIonizable, NegIonizable, Aromatic, Hydrophobe), if you want a fuzzier description of your molecules.

Bibliography

Carhart, R. E., Smith, D. H., & Venkataraghavan, R. (1985). Atom pairs as molecular features in structure-activity studies: definition and applications. Journal of Chemical Information and Computer Sciences, 25(2), 64-73.

Kearsley, S. K., Sallamack, S., Fluder, E. M., Andose, J. D., Mosley, R. T., & Sheridan, R. P. (1996). Chemical similarity using physiochemical property descriptors. Journal of Chemical Information and Computer Sciences, 36(1), 118-127.

Faulon, J. L., Visco, D. P., & Pophale, R. S. (2003). The signature molecular descriptor. 1. Using extended valence sequences in QSAR and QSPR studies. Journal of chemical information and computer sciences, 43(3), 707-720.

You can’t perform that action at this time.