Saagar-A New, Extensible Set of Molecular Substructures for QSAR/QSPR and Read-Across Predictions
Molecular structure-based predictive models provide a proven alternative to costly and inefficient animal testing. However, due to a lack of interpretability of predictive models built with abstract molecular descriptors they have earned the notoriety of being black boxes. Interpretable models require interpretable descriptors to provide chemistry-backed predictive reasoning and facilitate intelligent molecular design. We developed a novel set of extensible chemistry-aware substructures, Saagar, to support interpretable predictive models and read-across protocols. Performance of Saagar in chemical characterization and search for structurally similar actives for read-across applications was compared with four publicly available fingerprint sets (MACCS (166), PubChem (881), ECFP4 (1024), ToxPrint (729)) in three benchmark sets (MUV, ULS, and Tox21) spanning ∼145 000 compounds and 78 molecular targets at 1%, 2%, 5%, and 10% false discovery rates. In 18 of the 20 comparisons, interpretable Saagar features performed better than the publicly available, but less interpretable and fixed-bit length, fingerprints. Examples are provided to show the enhanced capability of Saagar in extracting compounds with higher scaffold similarity. Saagar features are interpretable and efficiently characterize diverse chemical collections, thus making them a better choice for building interpretable predictive in silico models and read-across protocols.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2021 |
|---|---|
| Erschienen: |
2021 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:34 |
|---|---|
| Enthalten in: |
Chemical research in toxicology - 34(2021), 2 vom: 15. Feb., Seite 634-640 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Sedykh, Alexander Y [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
Anthraquinones |
|---|
| Anmerkungen: |
Date Completed 23.09.2021 Date Revised 23.09.2021 published: Print-Electronic Citation Status MEDLINE |
|---|
| doi: |
10.1021/acs.chemrestox.0c00464 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM319205509 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM319205509 | ||
| 003 | DE-627 | ||
| 005 | 20231225170943.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231225s2021 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1021/acs.chemrestox.0c00464 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1063.xml |
| 035 | |a (DE-627)NLM319205509 | ||
| 035 | |a (NLM)33356152 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Sedykh, Alexander Y |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Saagar-A New, Extensible Set of Molecular Substructures for QSAR/QSPR and Read-Across Predictions |
| 264 | 1 | |c 2021 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 23.09.2021 | ||
| 500 | |a Date Revised 23.09.2021 | ||
| 500 | |a published: Print-Electronic | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a Molecular structure-based predictive models provide a proven alternative to costly and inefficient animal testing. However, due to a lack of interpretability of predictive models built with abstract molecular descriptors they have earned the notoriety of being black boxes. Interpretable models require interpretable descriptors to provide chemistry-backed predictive reasoning and facilitate intelligent molecular design. We developed a novel set of extensible chemistry-aware substructures, Saagar, to support interpretable predictive models and read-across protocols. Performance of Saagar in chemical characterization and search for structurally similar actives for read-across applications was compared with four publicly available fingerprint sets (MACCS (166), PubChem (881), ECFP4 (1024), ToxPrint (729)) in three benchmark sets (MUV, ULS, and Tox21) spanning ∼145 000 compounds and 78 molecular targets at 1%, 2%, 5%, and 10% false discovery rates. In 18 of the 20 comparisons, interpretable Saagar features performed better than the publicly available, but less interpretable and fixed-bit length, fingerprints. Examples are provided to show the enhanced capability of Saagar in extracting compounds with higher scaffold similarity. Saagar features are interpretable and efficiently characterize diverse chemical collections, thus making them a better choice for building interpretable predictive in silico models and read-across protocols | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a Research Support, N.I.H., Extramural | |
| 650 | 7 | |a Anthraquinones |2 NLM | |
| 700 | 1 | |a Shah, Ruchir R |e verfasserin |4 aut | |
| 700 | 1 | |a Kleinstreuer, Nicole C |e verfasserin |4 aut | |
| 700 | 1 | |a Auerbach, Scott S |e verfasserin |4 aut | |
| 700 | 1 | |a Gombar, Vijay K |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Chemical research in toxicology |d 1989 |g 34(2021), 2 vom: 15. Feb., Seite 634-640 |w (DE-627)NLM01260321X |x 1520-5010 |7 nnns |
| 773 | 1 | 8 | |g volume:34 |g year:2021 |g number:2 |g day:15 |g month:02 |g pages:634-640 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1021/acs.chemrestox.0c00464 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 34 |j 2021 |e 2 |b 15 |c 02 |h 634-640 | ||