2020/08/30 by María Virginia Sabando, Pavol Ulbrich, Sabando, María Virginia +15
Biochemistry, Genetics and Molecular Biology · Computer Science · Materials Science · #Cell Image Analysis Techniques #Computational Drug Discovery Methods #FOS: Computer and information sciences #Graphics (cs.GR) #Machine Learning in Materials Science
paper · pdf · doi:10.48550/arxiv.2008.13150
openalex publication_date 2020/08/30 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28
In the modern drug discovery process, medicinal chemists deal with the\ncomplexity of analysis of large ensembles of candidate molecules. Computational\ntools, such as dimensionality reduction (DR) and classification, are commonly\nused to efficiently process the multidimensional space of features. These\nunderlying calculations often hinder interpretability of results and prevent\nexperts from assessing the impact of individual molecular features on the\nresulting representations. To provide a solution for scrutinizing such complex\ndata, we introduce ChemVA, an interactive application for the visual\nexploration of large molecular ensembles and their features. Our tool consists\nof multiple coordinated views: Hexagonal view, Detail view, 3D view, Table\nview, and a newly proposed Difference view designed for the comparison of DR\nprojections. These views display DR projections combined with biological\nactivity, selected molecular features, and confidence scores for each of these\nprojections. This conjunction of views allows the user to drill down through\nthe dataset and to efficiently select candidate compounds. Our approach was\nevaluated on two case studies of finding structurally similar ligands with\nsimilar binding affinity to a target protein, as well as on an external\nqualitative evaluation. The results suggest that our system allows effective\nvisual inspection and comparison of different high-dimensional molecular\nrepresentations. Furthermore, ChemVA assists in the identification of candidate\ncompounds while providing information on the certainty behind different\nmolecular representations.\n