Introducing a Chemically Intuitive Core-Substituent Fingerprint Designed to Explore Structural Requirements for Effective Similarity Searching and Machine Learning

Abstract

Fingerprint (FP) representations of chemical structure continue to be one of the most widely used types of molecular descriptors in chemoinformatics and computational medicinal chemistry. One often distinguishes between two- and three-dimensional (2D and 3D) FPs depending on whether they are derived from molecular graphs or conformations, respectively. Primary application areas for FPs include similarity searching and compound classification via machine learning, especially for hit identification. For these applications, 2D FPs are particularly popular, given their robustness and for the most part comparable (or better) performance to 3D FPs. While a variety of FP prototypes has been designed and evaluated during earlier times of chemoinformatics research, new developments have been rare over the past decade. At least in part, this has been due to the situation that topological (atom environment) FPs derived from molecular graphs have evolved as a gold standard in the field. We were interested in exploring the question of whether the amount of structural information captured by state-of-the-art 2D FPs is indeed required for effective similarity searching and compound classification or whether accounting for fewer structural features might be sufficient. Therefore, pursuing a “structural minimalist” approach, we designed and implemented a new 2D FP based upon ring and substituent fragments obtained by systematically decomposing large numbers of compounds from medicinal chemistry. The resulting FP termed core-substituent FP (CSFP) captures much smaller numbers of structural features than state-of-the-art 2D FPs. However, CSFP achieves high performance in similarity searching and machine learning, demonstrating that less structural information is required for establishing molecular similarity relationships than is often believed. Given its high performance and chemical tangibility, CSFP is also relevant for practical applications in medicinal chemistry.

CIRCE: Web-Based Platform for the Prediction of Cannabinoid Receptor Ligands Using Explainable Machine Learning

“…Each structural fragment was assigned to a single bit position. For further details, the interested reader is referred to the original work …”

“…For further details, the interested reader is referred to the original work. 21 Machine Learning Models. Classification models were built by using the RF algorithm, with a default number of trees (n_estimator) equal to 400, 53 implemented using the scikitlearn python package.…”

“…Specifically, four cooperating classification models were generated for predicting and rationalizing molecular determinants driving selective ligand binding to CB 1 R and CB 2 R. Each classification model was independently derived based on different sets of training data carefully curated from the ChEMBL database (release 31) . These compound pools contain a wealth of chemical information along with high-quality experimental data for binding to CB 1 R and CB 2 R. A new substructure-based core-substituent fingerprint (CSFP) was used to encode the structural information, and SHAP values − were computed to explain individual predictions. Eventually, the four models were assembled to build a multilayer classifier, which is freely accessible through a web platform designated Cannabinoid Iterative Revaluation for Classification and Explanation (CIRCE) that is provided with a user-friendly graphical interface. , CIRCE returns predictions on demand and instantly provides a detailed portable report of prediction outcomes.…”

TISBE: A Public Web Platform for the Consensus-Based Explainable Prediction of Developmental Toxicity

“…47 Here, a recently proposed selective FP composed of 1000 bits, better known as CSFP, was used. 50 In particular, CSFP 63 represents only 2D structures, thus limiting the representation redundancy. 50 To further reduce dimension, 420 identical null bits were removed when generating the CSFP from our pool of Dev Tox data.…”

Untitled

“…The latter results from special pruning methods or removal processes such as FRED / S-KEYS (fast random removal of descriptor/substructure keys). The newly announced hologram-QSAR (H-QSAR) technology relies on calculating the occurrence of substructured pathways for specific functional groups 57,58 . It no longer relies on a predefined list of substructure motifs 59,60 .…”