EMBER—Embedding Multiple Molecular Fingerprints for Virtual Screening
In recent years, the debate in the field of applications of Deep Learning to Virtual Screening has focused on the use of neural embeddings with respect to classical descriptors in order to encode both structural and physical properties of ligands and/or targets. The attention on embeddings with the...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-02-01
|
| Series: | International Journal of Molecular Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1422-0067/23/4/2156 |
| _version_ | 1797479307377901568 |
|---|---|
| author | Isabella Mendolia Salvatore Contino Giada De Simone Ugo Perricone Roberto Pirrone |
| author_facet | Isabella Mendolia Salvatore Contino Giada De Simone Ugo Perricone Roberto Pirrone |
| author_sort | Isabella Mendolia |
| collection | DOAJ |
| description | In recent years, the debate in the field of applications of Deep Learning to Virtual Screening has focused on the use of neural embeddings with respect to classical descriptors in order to encode both structural and physical properties of ligands and/or targets. The attention on embeddings with the increasing use of Graph Neural Networks aimed at overcoming molecular fingerprints that are short range embeddings for atomic neighborhoods. Here, we present EMBER, a novel molecular embedding made by seven molecular fingerprints arranged as different “spectra” to describe the same molecule, and we prove its effectiveness by using deep convolutional architecture that assesses ligands’ bioactivity on a data set containing twenty protein kinases with similar binding sites to CDK1. The data set itself is presented, and the architecture is explained in detail along with its training procedure. We report experimental results and an explainability analysis to assess the contribution of each fingerprint to different targets. |
| first_indexed | 2024-03-09T21:43:59Z |
| format | Article |
| id | doaj.art-5123c6ea136e4c829fbbfdc775ac2760 |
| institution | Directory Open Access Journal |
| issn | 1661-6596 1422-0067 |
| language | English |
| last_indexed | 2024-03-09T21:43:59Z |
| publishDate | 2022-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | International Journal of Molecular Sciences |
| spelling | doaj.art-5123c6ea136e4c829fbbfdc775ac27602023-11-23T20:20:53ZengMDPI AGInternational Journal of Molecular Sciences1661-65961422-00672022-02-01234215610.3390/ijms23042156EMBER—Embedding Multiple Molecular Fingerprints for Virtual ScreeningIsabella Mendolia0Salvatore Contino1Giada De Simone2Ugo Perricone3Roberto Pirrone4Dipartimento di Ingegneria, Università degli Studi di Palermo, 90133 Palermo, ItalyDipartimento di Ingegneria, Università degli Studi di Palermo, 90133 Palermo, ItalyMolecular Informatics Group, Fondazione Ri.MED, 90133 Palermo, ItalyMolecular Informatics Group, Fondazione Ri.MED, 90133 Palermo, ItalyDipartimento di Ingegneria, Università degli Studi di Palermo, 90133 Palermo, ItalyIn recent years, the debate in the field of applications of Deep Learning to Virtual Screening has focused on the use of neural embeddings with respect to classical descriptors in order to encode both structural and physical properties of ligands and/or targets. The attention on embeddings with the increasing use of Graph Neural Networks aimed at overcoming molecular fingerprints that are short range embeddings for atomic neighborhoods. Here, we present EMBER, a novel molecular embedding made by seven molecular fingerprints arranged as different “spectra” to describe the same molecule, and we prove its effectiveness by using deep convolutional architecture that assesses ligands’ bioactivity on a data set containing twenty protein kinases with similar binding sites to CDK1. The data set itself is presented, and the architecture is explained in detail along with its training procedure. We report experimental results and an explainability analysis to assess the contribution of each fingerprint to different targets.https://www.mdpi.com/1422-0067/23/4/2156deep learningdrug designvirtual screeningembedding |
| spellingShingle | Isabella Mendolia Salvatore Contino Giada De Simone Ugo Perricone Roberto Pirrone EMBER—Embedding Multiple Molecular Fingerprints for Virtual Screening International Journal of Molecular Sciences deep learning drug design virtual screening embedding |
| title | EMBER—Embedding Multiple Molecular Fingerprints for Virtual Screening |
| title_full | EMBER—Embedding Multiple Molecular Fingerprints for Virtual Screening |
| title_fullStr | EMBER—Embedding Multiple Molecular Fingerprints for Virtual Screening |
| title_full_unstemmed | EMBER—Embedding Multiple Molecular Fingerprints for Virtual Screening |
| title_short | EMBER—Embedding Multiple Molecular Fingerprints for Virtual Screening |
| title_sort | ember embedding multiple molecular fingerprints for virtual screening |
| topic | deep learning drug design virtual screening embedding |
| url | https://www.mdpi.com/1422-0067/23/4/2156 |
| work_keys_str_mv | AT isabellamendolia emberembeddingmultiplemolecularfingerprintsforvirtualscreening AT salvatorecontino emberembeddingmultiplemolecularfingerprintsforvirtualscreening AT giadadesimone emberembeddingmultiplemolecularfingerprintsforvirtualscreening AT ugoperricone emberembeddingmultiplemolecularfingerprintsforvirtualscreening AT robertopirrone emberembeddingmultiplemolecularfingerprintsforvirtualscreening |