[Rdkit-discuss] chemfp 3.5.1 and ChEMBL 27 FPB distributions

[Andrew D. <da...@da...>]

Hi all,

  I've just released chemfp 3.5.1 with support for "licensed
FPB files". These are fingerprint datasets which can be used
under the terms of chemfp's base license agreement even without
a chemfp license key or source code distribution.

As the first (and so far only) data set, I've converted the
RDKit Morgan fingerprints from the ChEMBL 27 release into FPB
format and made it available from https://chemfp.com/datasets/ .

If you are on a Linux-based OS and RDKit is already installed
then here are the steps to get started:

1) Install a pre-compiled version of chemfp for Linux:

  python -m pip install chemfp -i https://chemfp.com/packages/

2) Download the ChEMBL data set in FPB format using one of
the following:

  wget https://chemfp.com/datasets/chembl_27.fpb.gz
        -or-
  curl -O https://chemfp.com/datasets/chembl_27.fpb.gz

3) (Optional but recommended) Uncompress it:

    gunzip chembl_27.fpb.gz

4) Do a similarity search, for example, with a query SMILES
or query file:

    simsearch --query c1ccccc1O chembl_27.fpb
    simsearch --queries queryfile.sdf chembl_27.fpb

5) View the ChEMBL license agreement and legal notices
included with the dataset.
 
    python -m chemfp fpb_text chembl_27.fpb


Chemfp is a Python package for cheminformatics fingerprints.

It can be used to:
  - generate fingerprints using the  RDKit, Open Babel,
      CDK, and OEChem/OEGraphSim toolkits;
  - extract pre-computed fingerprints from SDF tag data;
  - do high-performance Tanimoto and Tversky similarity search;
  - integrate fingerprints and search results with NumPy/SciPy.
  - ... and much more!

It includes an extensive and well-documented Python API for
working with fingerprints and a set of command-line tools
for fingerprint generation, conversion, and similarity search.

Chemfp natively supports two fingerprint file formats. The
FPS format is a text format which is easy to read and write.
The FPB format is a binary format which is quick to load.

For more information see https://chemfp.com/ .

License keys for academic use are available at no cost.

Best regards,

				Andrew
				da...@da...