Multi-instance Learning for Structure-Activity Modeling for Molecular Properties

Zankov, Dmitry; Shevelev, M. D.; Nikonenko, Alexandra V.; Polishchuk, Pavel; Rakhimbekova, Assima; Madzhidov, Timur

doi:10.1007/978-3-030-39575-9_7

Cited by 2 publications

(6 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…26 Recently, we demonstrated the applicability of unsupervised 27 and supervised clustering-based MI approaches to bioactivity predictions on several data sets. 28 However, a proper comparison of MI learning approaches to conventional ones has not been made so far.…”

Section: ■ Introductionmentioning

confidence: 99%

“…Only a few studies with the application of MI learning to predict the bioactivity of the compounds have been published so far in mathematics and bioinformatics journals. ,− Moreover, recently proposed deep learning-based multi-instance approaches have not been used in the chemistry domain except in our recent work . Recently, we demonstrated the applicability of unsupervised and supervised clustering-based MI approaches to bioactivity predictions on several data sets . However, a proper comparison of MI learning approaches to conventional ones has not been made so far.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

QSAR Modeling Based on Conformation Ensembles Using a Multi-Instance Learning Approach

Zankov

Matveieva

Nikonenko

et al. 2021

J. Chem. Inf. Model.

Self Cite

View full text Add to dashboard Cite

Modern QSAR approaches have wide practical applications in drug discovery for designing potentially bioactive molecules. If such models are based on the use of 2D descriptors, important information contained in the spatial structures of molecules is lost. The major problem in constructing models using 3D descriptors is the choice of a putative bioactive conformation, which affects the predictive performance. The multi-instance (MI) learning approach considering multiple conformations in model training could be a reasonable solution to the above problem. In this study, we implemented several multi-instance algorithms, both conventional and based on deep learning, and investigated their performance. We compared the performance of MI-QSAR models with those based on the classical single-instance QSAR (SI-QSAR) approach in which each molecule is encoded by either 2D descriptors computed for the corresponding molecular graph or 3D descriptors issued for a single lowest energy conformation. The calculations were carried out on 175 data sets extracted from the ChEMBL23 database. It is demonstrated that (i) MI-QSAR outperforms SI-QSAR in numerous cases and (ii) MI algorithms can automatically identify plausible bioactive conformations.

show abstract

Section: ■ Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

QSAR Modeling Based on Conformation Ensembles Using a Multi-Instance Learning Approach

Zankov

Matveieva

Nikonenko

et al. 2021

J. Chem. Inf. Model.

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, it was outperformed by conventional 2D models in a larger scale testing on 163 datasets extracted from ChEMBL which contained preferably achiral molecules: 2D models were better in 124 cases, whereas in other cases both models demonstrated accuracy close to random. Afterward, we extended our studies 17,18 and performed a large-scale benchmark of single-instance and multiinstance regression models for the prediction of the biological activity of molecules on an updated 175 datasets from the ChEMBL database. 3D multi-conformation models outperformed 3D single-conformation models in 98% of cases (average R 2 = 0.524 vs. 0.024) and conventional 2D models in 70% of cases 17 (average R 2 = 0.524 vs. 0.464).…”

Section: Multi-instance Learning Applicationsmentioning

confidence: 99%

“…Afterward, we extended our studies 17,18 and performed a large‐scale benchmark of single‐instance and multi‐instance regression models for the prediction of the biological activity of molecules on an updated 175 datasets from the ChEMBL database. 3D multi‐conformation models outperformed 3D single‐conformation models in 98% of cases (average R 2 = 0.524 vs. 0.024) and conventional 2D models in 70% of cases 17 (average R 2 = 0.524 vs. 0.464).…”

Section: Multi‐instance Learning Applicationsmentioning

confidence: 99%

“…Since the seminal paper of Dietterich et al, 2 numerous MIL algorithms have been developed and applied in various domains, such as drug discovery (pharmacy), classification of text documents (information retrieval), classification of images (computer vision), speaker identification (signal processing), bankruptcy prediction (economy), and so on 9–11 . However, MIL still has not become a popular approach in chemoinformatics and only a few studies on its application to structure–property modeling have been reported so far 2,5,12–21 . In bioinformatics, MIL has attracted significantly more attention, because of a large number of tasks 22–35 perfectly fitting the MIL framework.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Chemical complexity challenge: Is multi‐instance machine learning a solution?

Zankov,

Madzhidov,

Varnek

et al. 2023

WIREs Comput Mol Sci

Self Cite

View full text Add to dashboard Cite

Molecules are complex dynamic objects that can exist in different molecular forms (conformations, tautomers, stereoisomers, protonation states, etc.) and often it is not known which molecular form is responsible for observed physicochemical and biological properties of a given molecule. This raises the problem of the selection of the correct molecular form for machine learning modeling of target properties. The same problem is common to biological molecules (RNA, DNA, proteins)—long sequences where only key segments, which often cannot be located precisely, are involved in biological functions. Multi‐instance machine learning (MIL) is an efficient approach for solving problems where objects under study cannot be uniquely represented by a single instance, but rather by a set of multiple alternative instances. Multi‐instance learning was formalized in 1997 and motivated by the problem of conformation selection in drug activity prediction tasks. Since then MIL has found a lot of applications in various domains, such as information retrieval, computer vision, signal processing, bankruptcy prediction, and so on. In the given review we describe the MIL framework and its applications to the tasks associated with ambiguity in the representation of small and biological molecules in chemoinformatics and bioinformatics. We have collected examples that demonstrate the advantages of MIL over the traditional single‐instance learning (SIL) approach. Special attention was paid to the ability of MIL models to identify key instances responsible for a modeling property.This article is categorized under: Data Science > Chemoinformatics Data Science > Artificial Intelligence/Machine Learning

show abstract

Multi-instance Learning for Structure-Activity Modeling for Molecular Properties

Cited by 2 publications

References 17 publications

QSAR Modeling Based on Conformation Ensembles Using a Multi-Instance Learning Approach

QSAR Modeling Based on Conformation Ensembles Using a Multi-Instance Learning Approach

Chemical complexity challenge: Is multi‐instance machine learning a solution?

Contact Info

Product

Resources

About