KCF-Convoy: efficient Python package to convert KEGG Chemical Function and Substructure fingerprints

Abstract

Motivation:In silico methodologies to assess pharmaceutical activity and toxicity are increasingly important in QSAR, and many chemical fingerprints have been developed to tackle this problem. Among them, KEGG Chemical Function and Substructure (KCF-S) has been shown to perform well in some pharmaceutical and metabolic studies. However, the software that generates KCF-S fingerprints has limited usability: the input file must be Molfile or SDF format, and the output is only a text file. Results:We established a new Python package, KCF-Convoy, to generate KCF format and KCF-S fingerprints from Molfile, SDF, SMILES, and InChI seamlessly. The obtained KCF-S was used in a number of supervised machine-learning methods to distinguish herbicides from other pesticides, and to find characteristic substructures in taxonomy groups.Availability: KCF-Convoy is implemented as a Python package freely available at https://github.com/KCF-Convoy and the user can use the package management system "pip" and also the Docker environment.

Quasi‐supervised Strategies for Compound‐protein Interaction Prediction

“…We used KCF-Convoy python package [43] to construct fingerprint vectors and calculated the similarity between these vectors using a weighted Tanimoto similarity. [43]…”

“…The dimension of these vectors equals the number of unique substructures listed in a database of substructures that can be extracted from the compounds. We used KCF‐Convoy python package [43] to construct fingerprint vectors and calculated the similarity between these vectors using a weighted Tanimoto similarity [43] …”

Molecular structure discovery for untargeted metabolomics using biotransformation rules and global molecular networking

“…BAM was implemented using two different code environments. First, PROXIMAL2 was applied and its results were analyzed using Python 3.9, PubChemPy (version 1.0.4) [19], BioServices (version 1.11.2) [20], scikit-learn (version 1.1.2) [21], NumPy (version 1.23.2) [22], NetworkX (version 2.5) [23], RDKit (version 2022.03.5) [24], KCF-Convoy (version 0.0.5) [25,26], Pandas (version 1.4.3) [27,28], Matplotlib (version 3.5.3) [29], and Seaborn (version 0.12.2) [30]. Second, GNN-SOM was applied using Python 3.10, RDKit (version 2022.03.5) [24], Py-Torch (version 1.12.1) [31,32], and PyTorch Geometric (version 2.1.0) [33].…”

Extending PROXIMAL to predict degradation pathways of phenolic compounds in the human gut microbiota

“…As noted above, reaction centres are defined as any specific substrate atom that aligns with a product atom of different KEGG atom type. Here, the classification of substrate and product atoms into different KEGG atom types were done with the package KCF-Convoy (Kotera et al, 2013;Sato et al, 2018). With this information and the atomic alignment (Step3), reaction centres were identified.…”