Large-scale prediction of activity cliffs using machine and deep learning methods of increasing complexity
Activity cliffs (AC) are formed by pairs of structural analogues that are active against the same target but have a large difference in potency. While much of our knowledge about ACs has originated from the analysis and comparison of compounds and activity data, several studies have reported AC pred...
Saved in:
| Published in: | Journal of cheminformatics Vol. 15; no. 1; pp. 4 - 11 |
|---|---|
| Main Authors: | , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
Cham
Springer International Publishing
07.01.2023
BioMed Central Ltd Springer Nature B.V BMC |
| Subjects: | |
| ISSN: | 1758-2946, 1758-2946 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | Activity cliffs (AC) are formed by pairs of structural analogues that are active against the same target but have a large difference in potency. While much of our knowledge about ACs has originated from the analysis and comparison of compounds and activity data, several studies have reported AC predictions over the past decade. Different from typical compound classification tasks, AC predictions must be carried out at the level of compound pairs representing ACs or nonACs. Most AC predictions reported so far have focused on individual methods or comparisons of two or three approaches and only investigated a few compound activity classes (from 2 to 10). Although promising prediction accuracy has been reported in most cases, different system set-ups, AC definitions, methods, and calculation conditions were used, precluding direct comparisons of these studies. Therefore, we have carried out a large-scale AC prediction campaign across 100 activity classes comparing machine learning methods of greatly varying complexity, ranging from pair-based nearest neighbor classifiers and decision tree or kernel methods to deep neural networks. The results of our systematic predictions revealed the level of accuracy that can be expected for AC predictions across many different compound classes. In addition, prediction accuracy did not scale with methodological complexity but was significantly influenced by memorization of compounds shared by different ACs or nonACs. In many instances, limited training data were sufficient for building accurate models using different methods and there was no detectable advantage of deep learning over simpler approaches for AC prediction. On a global scale, support vector machine models performed best, by only small margins compared to others including simple nearest neighbor classifiers.
Graphical Abstract |
|---|---|
| AbstractList | Activity cliffs (AC) are formed by pairs of structural analogues that are active against the same target but have a large difference in potency. While much of our knowledge about ACs has originated from the analysis and comparison of compounds and activity data, several studies have reported AC predictions over the past decade. Different from typical compound classification tasks, AC predictions must be carried out at the level of compound pairs representing ACs or nonACs. Most AC predictions reported so far have focused on individual methods or comparisons of two or three approaches and only investigated a few compound activity classes (from 2 to 10). Although promising prediction accuracy has been reported in most cases, different system set-ups, AC definitions, methods, and calculation conditions were used, precluding direct comparisons of these studies. Therefore, we have carried out a large-scale AC prediction campaign across 100 activity classes comparing machine learning methods of greatly varying complexity, ranging from pair-based nearest neighbor classifiers and decision tree or kernel methods to deep neural networks. The results of our systematic predictions revealed the level of accuracy that can be expected for AC predictions across many different compound classes. In addition, prediction accuracy did not scale with methodological complexity but was significantly influenced by memorization of compounds shared by different ACs or nonACs. In many instances, limited training data were sufficient for building accurate models using different methods and there was no detectable advantage of deep learning over simpler approaches for AC prediction. On a global scale, support vector machine models performed best, by only small margins compared to others including simple nearest neighbor classifiers. Graphical Keywords: Activity cliff, Machine learning, Deep learning, Compound pair-based prediction, Large-scale analysis Activity cliffs (AC) are formed by pairs of structural analogues that are active against the same target but have a large difference in potency. While much of our knowledge about ACs has originated from the analysis and comparison of compounds and activity data, several studies have reported AC predictions over the past decade. Different from typical compound classification tasks, AC predictions must be carried out at the level of compound pairs representing ACs or nonACs. Most AC predictions reported so far have focused on individual methods or comparisons of two or three approaches and only investigated a few compound activity classes (from 2 to 10). Although promising prediction accuracy has been reported in most cases, different system set-ups, AC definitions, methods, and calculation conditions were used, precluding direct comparisons of these studies. Therefore, we have carried out a large-scale AC prediction campaign across 100 activity classes comparing machine learning methods of greatly varying complexity, ranging from pair-based nearest neighbor classifiers and decision tree or kernel methods to deep neural networks. The results of our systematic predictions revealed the level of accuracy that can be expected for AC predictions across many different compound classes. In addition, prediction accuracy did not scale with methodological complexity but was significantly influenced by memorization of compounds shared by different ACs or nonACs. In many instances, limited training data were sufficient for building accurate models using different methods and there was no detectable advantage of deep learning over simpler approaches for AC prediction. On a global scale, support vector machine models performed best, by only small margins compared to others including simple nearest neighbor classifiers. Activity cliffs (AC) are formed by pairs of structural analogues that are active against the same target but have a large difference in potency. While much of our knowledge about ACs has originated from the analysis and comparison of compounds and activity data, several studies have reported AC predictions over the past decade. Different from typical compound classification tasks, AC predictions must be carried out at the level of compound pairs representing ACs or nonACs. Most AC predictions reported so far have focused on individual methods or comparisons of two or three approaches and only investigated a few compound activity classes (from 2 to 10). Although promising prediction accuracy has been reported in most cases, different system set-ups, AC definitions, methods, and calculation conditions were used, precluding direct comparisons of these studies. Therefore, we have carried out a large-scale AC prediction campaign across 100 activity classes comparing machine learning methods of greatly varying complexity, ranging from pair-based nearest neighbor classifiers and decision tree or kernel methods to deep neural networks. The results of our systematic predictions revealed the level of accuracy that can be expected for AC predictions across many different compound classes. In addition, prediction accuracy did not scale with methodological complexity but was significantly influenced by memorization of compounds shared by different ACs or nonACs. In many instances, limited training data were sufficient for building accurate models using different methods and there was no detectable advantage of deep learning over simpler approaches for AC prediction. On a global scale, support vector machine models performed best, by only small margins compared to others including simple nearest neighbor classifiers. Graphical Abstract Activity cliffs (AC) are formed by pairs of structural analogues that are active against the same target but have a large difference in potency. While much of our knowledge about ACs has originated from the analysis and comparison of compounds and activity data, several studies have reported AC predictions over the past decade. Different from typical compound classification tasks, AC predictions must be carried out at the level of compound pairs representing ACs or nonACs. Most AC predictions reported so far have focused on individual methods or comparisons of two or three approaches and only investigated a few compound activity classes (from 2 to 10). Although promising prediction accuracy has been reported in most cases, different system set-ups, AC definitions, methods, and calculation conditions were used, precluding direct comparisons of these studies. Therefore, we have carried out a large-scale AC prediction campaign across 100 activity classes comparing machine learning methods of greatly varying complexity, ranging from pair-based nearest neighbor classifiers and decision tree or kernel methods to deep neural networks. The results of our systematic predictions revealed the level of accuracy that can be expected for AC predictions across many different compound classes. In addition, prediction accuracy did not scale with methodological complexity but was significantly influenced by memorization of compounds shared by different ACs or nonACs. In many instances, limited training data were sufficient for building accurate models using different methods and there was no detectable advantage of deep learning over simpler approaches for AC prediction. On a global scale, support vector machine models performed best, by only small margins compared to others including simple nearest neighbor classifiers.Activity cliffs (AC) are formed by pairs of structural analogues that are active against the same target but have a large difference in potency. While much of our knowledge about ACs has originated from the analysis and comparison of compounds and activity data, several studies have reported AC predictions over the past decade. Different from typical compound classification tasks, AC predictions must be carried out at the level of compound pairs representing ACs or nonACs. Most AC predictions reported so far have focused on individual methods or comparisons of two or three approaches and only investigated a few compound activity classes (from 2 to 10). Although promising prediction accuracy has been reported in most cases, different system set-ups, AC definitions, methods, and calculation conditions were used, precluding direct comparisons of these studies. Therefore, we have carried out a large-scale AC prediction campaign across 100 activity classes comparing machine learning methods of greatly varying complexity, ranging from pair-based nearest neighbor classifiers and decision tree or kernel methods to deep neural networks. The results of our systematic predictions revealed the level of accuracy that can be expected for AC predictions across many different compound classes. In addition, prediction accuracy did not scale with methodological complexity but was significantly influenced by memorization of compounds shared by different ACs or nonACs. In many instances, limited training data were sufficient for building accurate models using different methods and there was no detectable advantage of deep learning over simpler approaches for AC prediction. On a global scale, support vector machine models performed best, by only small margins compared to others including simple nearest neighbor classifiers. Abstract Activity cliffs (AC) are formed by pairs of structural analogues that are active against the same target but have a large difference in potency. While much of our knowledge about ACs has originated from the analysis and comparison of compounds and activity data, several studies have reported AC predictions over the past decade. Different from typical compound classification tasks, AC predictions must be carried out at the level of compound pairs representing ACs or nonACs. Most AC predictions reported so far have focused on individual methods or comparisons of two or three approaches and only investigated a few compound activity classes (from 2 to 10). Although promising prediction accuracy has been reported in most cases, different system set-ups, AC definitions, methods, and calculation conditions were used, precluding direct comparisons of these studies. Therefore, we have carried out a large-scale AC prediction campaign across 100 activity classes comparing machine learning methods of greatly varying complexity, ranging from pair-based nearest neighbor classifiers and decision tree or kernel methods to deep neural networks. The results of our systematic predictions revealed the level of accuracy that can be expected for AC predictions across many different compound classes. In addition, prediction accuracy did not scale with methodological complexity but was significantly influenced by memorization of compounds shared by different ACs or nonACs. In many instances, limited training data were sufficient for building accurate models using different methods and there was no detectable advantage of deep learning over simpler approaches for AC prediction. On a global scale, support vector machine models performed best, by only small margins compared to others including simple nearest neighbor classifiers. Graphical Abstract Activity cliffs (AC) are formed by pairs of structural analogues that are active against the same target but have a large difference in potency. While much of our knowledge about ACs has originated from the analysis and comparison of compounds and activity data, several studies have reported AC predictions over the past decade. Different from typical compound classification tasks, AC predictions must be carried out at the level of compound pairs representing ACs or nonACs. Most AC predictions reported so far have focused on individual methods or comparisons of two or three approaches and only investigated a few compound activity classes (from 2 to 10). Although promising prediction accuracy has been reported in most cases, different system set-ups, AC definitions, methods, and calculation conditions were used, precluding direct comparisons of these studies. Therefore, we have carried out a large-scale AC prediction campaign across 100 activity classes comparing machine learning methods of greatly varying complexity, ranging from pair-based nearest neighbor classifiers and decision tree or kernel methods to deep neural networks. The results of our systematic predictions revealed the level of accuracy that can be expected for AC predictions across many different compound classes. In addition, prediction accuracy did not scale with methodological complexity but was significantly influenced by memorization of compounds shared by different ACs or nonACs. In many instances, limited training data were sufficient for building accurate models using different methods and there was no detectable advantage of deep learning over simpler approaches for AC prediction. On a global scale, support vector machine models performed best, by only small margins compared to others including simple nearest neighbor classifiers. |
| ArticleNumber | 4 |
| Audience | Academic |
| Author | Miyao, Tomoyuki Tamura, Shunsuke Bajorath, Jürgen |
| Author_xml | – sequence: 1 givenname: Shunsuke surname: Tamura fullname: Tamura, Shunsuke organization: Department of Life Science Informatics, B-IT, LIMES Program Unit Chemical Biology and Medicinal Chemistry, Rheinische Friedrich-Wilhelms-Universität, Graduate School of Science and Technology, Nara Institute of Science and Technology – sequence: 2 givenname: Tomoyuki surname: Miyao fullname: Miyao, Tomoyuki email: miyao@dsc.naist.jp organization: Graduate School of Science and Technology, Nara Institute of Science and Technology, Data Science Center, Nara Institute of Science and Technology – sequence: 3 givenname: Jürgen surname: Bajorath fullname: Bajorath, Jürgen email: bajorath@bit-uni.bonn.de organization: Department of Life Science Informatics, B-IT, LIMES Program Unit Chemical Biology and Medicinal Chemistry, Rheinische Friedrich-Wilhelms-Universität |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/36611204$$D View this record in MEDLINE/PubMed |
| BookMark | eNp9kktr3DAUhU1JaR7tH-iiGLpJF071sCVrUwihj4GBQh9rIUtXHg0eaSrZIfn31XjSJBNK0EJC9ztHl6tzWhz54KEo3mJ0gXHLPiZMKcEVIqRCiHFW8RfFCeZNWxFRs6NH5-PiNKV1hhqO-KvimDKGMUH1SbFeqthDlbQaoNxGME6PLvgy2FLl07Ubb0s9OGtTOSXn-3Kj9Mp5KJU3pQHYlgOo6OcKjKtg0k7qvI6gZl6HzXaAm-zzunhp1ZDgzd1-Vvz-8vnX1bdq-f3r4upyWWne4LECww3tdCuM4lp0NHdNiLWMtp2gDKhqhBJWI2GIQhhbTjRrWqCtQYAaoulZsdj7mqDWchvdRsVbGZST80WIvVRxdHoACZYRroGS1kLdai6Y4EYxBl2njTFd9vq099pO3QaMBj9GNRyYHla8W8k-XEvRkgbVKBuc3xnE8GeCNMqNSxqGQXkIU5KEMyxaxEWT0fdP0HWYos-jminGSF2LB6rPPyadtyG_q3em8pJT0tQ1n6mL_1B5Gdg4nWNkXb4_EHw4EGRmhJuxV1NKcvHzxyH77vFQ7qfxL1UZIHtAx5BSBHuPYCR30ZX76MocXTlHV_Isap-ItBvVLoy5dTc8L6V7acrv-B7iw-SeUf0FhXYCcw |
| CitedBy_id | crossref_primary_10_1007_s10822_024_00566_0 crossref_primary_10_1186_s13321_025_01006_3 crossref_primary_10_1039_D3ME00189J crossref_primary_10_1021_acs_jcim_5c00484 crossref_primary_10_1038_s41597_024_03371_4 crossref_primary_10_1186_s13321_023_00733_9 crossref_primary_10_3390_ph16091259 crossref_primary_10_1016_j_ejmech_2025_117269 crossref_primary_10_1289_EHP14001 crossref_primary_10_1021_acs_jcim_5c00506 |
| Cites_doi | 10.1186/s13321-020-0414-z 10.1093/nar/gky1075 10.1007/s10822-022-00449-2 10.1214/aos/1013203451 10.1021/ci900450m 10.1021/ci3001138 10.1016/0005-2795(75)90109-9 10.3390/molecules26164916 10.1016/j.ailsci.2021.100022 10.1021/jm201706b 10.1021/acs.jcim.1c01163 10.1021/acs.jcim.6b00359 10.1007/978-1-4757-3264-1 10.1021/ci300047k 10.1002/minf.202000103 10.1021/ci060117s 10.1039/d2dd00077f 10.1007/s10822-021-00380-y 10.1021/ci300149n 10.1021/jm200026b 10.1021/acsomega.9b02221 10.1021/ci100050t 10.1021/ci300306a 10.1021/ci500742b 10.48550/arxiv.1412.6980 10.1021/acs.jcim.2c00327 10.4155/fmc-2018-0299 10.1016/j.neunet.2005.07.009 10.1023/a:1010933404324 10.1021/jm401120g 10.1145/3292500.3330701 10.1109/ICPR.2010.764 10.26434/chemrxiv-2022-mfq52-v3 |
| ContentType | Journal Article |
| Copyright | The Author(s) 2023 2023. The Author(s). COPYRIGHT 2023 BioMed Central Ltd. The Author(s) 2023. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
| Copyright_xml | – notice: The Author(s) 2023 – notice: 2023. The Author(s). – notice: COPYRIGHT 2023 BioMed Central Ltd. – notice: The Author(s) 2023. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
| DBID | C6C AAYXX CITATION NPM ISR 3V. 7QO 7X7 7XB 8AO 8FD 8FE 8FG 8FH 8FI 8FJ 8FK ABJCF ABUWG AEUYN AFKRA ARAPS AZQEC BBNVY BENPR BGLVJ BHPHI CCPQU D1I DWQXO FR3 FYUFA GHDGH GNUQQ HCIFZ K9. KB. LK8 M0S M7P P5Z P62 P64 PDBOC PHGZM PHGZT PIMPY PJZUB PKEHL PPXIY PQEST PQGLB PQQKQ PQUKI 7X8 5PM DOA |
| DOI | 10.1186/s13321-022-00676-7 |
| DatabaseName | Springer Nature OA Free Journals CrossRef PubMed Gale In Context: Science ProQuest Central (Corporate) Biotechnology Research Abstracts Health Medical collection ProQuest Central (purchase pre-March 2016) ProQuest Pharma Collection Technology Research Database ProQuest SciTech Collection ProQuest Technology Collection ProQuest Natural Science Collection Hospital Premium Collection Hospital Premium Collection (Alumni Edition) ProQuest Central (Alumni) (purchase pre-March 2016) SciTech Premium Collection ProQuest Central (Alumni) ProQuest One Sustainability (subscription) ProQuest Central UK/Ireland Health Research Premium Collection ProQuest Central Essentials Biological Science Collection ProQuest Central Technology collection Natural Science Collection ProQuest One Community College ProQuest Materials Science Collection ProQuest Central Engineering Research Database Health Research Premium Collection Health Research Premium Collection (Alumni) ProQuest Central Student SciTech Premium Collection ProQuest Health & Medical Complete (Alumni) Materials Science Database ProQuest Biological Science Collection Health & Medical Collection (Alumni Edition) Biological Science Database Advanced Technologies & Aerospace Database ProQuest Advanced Technologies & Aerospace Collection Biotechnology and BioEngineering Abstracts Materials Science Collection ProQuest Central Premium ProQuest One Academic (New) ProQuest Publicly Available Content Database ProQuest Health & Medical Research Collection ProQuest One Academic Middle East (New) ProQuest One Health & Nursing ProQuest One Academic Eastern Edition (DO NOT USE) ProQuest One Applied & Life Sciences ProQuest One Academic (retired) ProQuest One Academic UKI Edition MEDLINE - Academic PubMed Central (Full Participant titles) DOAJ - Directory of Open Access Journals |
| DatabaseTitle | CrossRef PubMed Publicly Available Content Database ProQuest Central Student Technology Collection Technology Research Database ProQuest One Academic Middle East (New) ProQuest Advanced Technologies & Aerospace Collection ProQuest Central Essentials Materials Science Collection ProQuest Health & Medical Complete (Alumni) ProQuest Central (Alumni Edition) SciTech Premium Collection ProQuest One Community College ProQuest One Health & Nursing ProQuest Natural Science Collection ProQuest Pharma Collection ProQuest Central ProQuest One Applied & Life Sciences ProQuest One Sustainability ProQuest Health & Medical Research Collection Health Research Premium Collection Biotechnology Research Abstracts Health and Medicine Complete (Alumni Edition) Natural Science Collection ProQuest Central Korea Biological Science Collection Materials Science Database ProQuest Central (New) ProQuest Materials Science Collection Advanced Technologies & Aerospace Collection ProQuest Biological Science Collection ProQuest One Academic Eastern Edition ProQuest Hospital Collection ProQuest Technology Collection Health Research Premium Collection (Alumni) Biological Science Database ProQuest SciTech Collection ProQuest Hospital Collection (Alumni) Biotechnology and BioEngineering Abstracts Advanced Technologies & Aerospace Database ProQuest Health & Medical Complete ProQuest One Academic UKI Edition Materials Science & Engineering Collection Engineering Research Database ProQuest One Academic ProQuest One Academic (New) ProQuest Central (Alumni) MEDLINE - Academic |
| DatabaseTitleList | CrossRef MEDLINE - Academic Publicly Available Content Database PubMed |
| Database_xml | – sequence: 1 dbid: DOA name: DOAJ - Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website – sequence: 2 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 3 dbid: PIMPY name: ProQuest Publicly Available Content Database url: http://search.proquest.com/publiccontent sourceTypes: Aggregation Database |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Chemistry |
| EISSN | 1758-2946 |
| EndPage | 11 |
| ExternalDocumentID | oai_doaj_org_article_ef627ce328fe48c79697da66ebbcdddb PMC9825040 A732544749 36611204 10_1186_s13321_022_00676_7 |
| Genre | Journal Article |
| GrantInformation_xml | – fundername: Japan Society for the Promotion of Science grantid: 22J12846 funderid: http://dx.doi.org/10.13039/501100001691 – fundername: Japan Society for the Promotion of Science grantid: 22J12846 – fundername: ; grantid: 22J12846 |
| GroupedDBID | -5F -5G -A0 -BR 0R~ 29K 2WC 3V. 4.4 40G 53G 5VS 7X7 8AO 8FE 8FG 8FH 8FI 8FJ AAFWJ AAJSJ AAKKN AAKPC ABDBF ABEEZ ABJCF ABUWG ACACY ACGFS ACIHN ACIWK ACPRK ACUHS ACULB ADBBV ADINQ ADRAZ ADUKV AEAQA AENEX AEUYN AFGXO AFKRA AFRAH AHBYD AHMBA AHYZX ALIPV ALMA_UNASSIGNED_HOLDINGS AMKLP AMTXH AOIJS ARAPS BAPOH BAWUL BBNVY BCNDV BENPR BFQNJ BGLVJ BHPHI BMC BPHCQ BVXVI C24 C6C CCPQU D-I D1I DIK E3Z EBLON EBS ESX F5P FRP FYUFA GROUPED_DOAJ GX1 HCIFZ HH5 HMCUK HYE IAO IGS IHR ISR ITC KB. KQ8 LK8 M48 M7P MK0 M~E O5R O5S OK1 P62 PDBOC PGMZT PIMPY PQQKQ PROAC RBZ RNS RPM RSV RVI SOJ SPH TR2 TUS U2A UKHRP AASML AAYXX AFPKN CITATION NPM 7QO 7XB 8FD 8FK AZQEC DWQXO FR3 GNUQQ K9. P64 PHGZM PHGZT PJZUB PKEHL PPXIY PQEST PQGLB PQUKI 7X8 PUEGO 5PM |
| ID | FETCH-LOGICAL-c751t-ed7d3bc89da7c9b300622ff638b936e3a59a9fc09d2a011f72c658e38d0e052c3 |
| IEDL.DBID | BENPR |
| ISICitedReferencesCount | 14 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000910831500001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1758-2946 |
| IngestDate | Tue Oct 14 19:08:27 EDT 2025 Tue Nov 04 02:06:50 EST 2025 Fri Sep 05 06:24:42 EDT 2025 Sat Oct 18 23:55:31 EDT 2025 Mon Oct 20 22:37:56 EDT 2025 Mon Oct 20 16:33:46 EDT 2025 Thu Oct 16 16:10:31 EDT 2025 Thu Jan 02 22:53:05 EST 2025 Sat Nov 29 05:55:23 EST 2025 Tue Nov 18 20:51:54 EST 2025 Fri Feb 21 02:46:15 EST 2025 |
| IsDoiOpenAccess | true |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 1 |
| Keywords | Deep learning Activity cliff Large-scale analysis Compound pair-based prediction Machine learning |
| Language | English |
| License | 2023. The Author(s). Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c751t-ed7d3bc89da7c9b300622ff638b936e3a59a9fc09d2a011f72c658e38d0e052c3 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23 |
| OpenAccessLink | https://www.proquest.com/docview/2761662449?pq-origsite=%requestingapplication% |
| PMID | 36611204 |
| PQID | 2761662449 |
| PQPubID | 54992 |
| PageCount | 11 |
| ParticipantIDs | doaj_primary_oai_doaj_org_article_ef627ce328fe48c79697da66ebbcdddb pubmedcentral_primary_oai_pubmedcentral_nih_gov_9825040 proquest_miscellaneous_2761980795 proquest_journals_2761662449 gale_infotracmisc_A732544749 gale_infotracacademiconefile_A732544749 gale_incontextgauss_ISR_A732544749 pubmed_primary_36611204 crossref_primary_10_1186_s13321_022_00676_7 crossref_citationtrail_10_1186_s13321_022_00676_7 springer_journals_10_1186_s13321_022_00676_7 |
| PublicationCentury | 2000 |
| PublicationDate | 2023-01-07 |
| PublicationDateYYYYMMDD | 2023-01-07 |
| PublicationDate_xml | – month: 01 year: 2023 text: 2023-01-07 day: 07 |
| PublicationDecade | 2020 |
| PublicationPlace | Cham |
| PublicationPlace_xml | – name: Cham – name: England – name: London |
| PublicationTitle | Journal of cheminformatics |
| PublicationTitleAbbrev | J Cheminform |
| PublicationTitleAlternate | J Cheminform |
| PublicationYear | 2023 |
| Publisher | Springer International Publishing BioMed Central Ltd Springer Nature B.V BMC |
| Publisher_xml | – name: Springer International Publishing – name: BioMed Central Ltd – name: Springer Nature B.V – name: BMC |
| References | Stumpfe, Hu, Dimova, Bajorath (CR2) 2014; 57 Stumpfe, Bajorath (CR20) 2012; 55 Maeda, Sato, Tamura, Miyao (CR35) 2022; 36 Hussain, Rea (CR18) 2010; 50 Stumpfe, Hu, Bajorath (CR3) 2019; 4 Hu, Hu, Vogt (CR17) 2012; 52 Breiman (CR31) 2001; 45 Park, Sung, Lee (CR12) 2022 Jiménez-Luna, Skalic, Weskamp (CR14) 2022; 62 CR15 CR37 Horvath, Marcou, Varnek (CR8) 2016; 56 CR33 Guha (CR4) 2012; 52 Hu, Stumpfe, Bajorath (CR21) 2019; 11 Matthews (CR38) 1975; 405 Heikamp, Hu, Yan, Bajorath (CR5) 2012; 52 Friedman (CR32) 2001 Tamura, Jasial, Miyao, Funatsu (CR7) 2021; 26 Rogers, Hahn (CR22) 2010; 50 Iqbal, Vogt, Bajorath (CR11) 2021 Tang, Kramer, Fang (CR36) 2020; 12 Husby, Bottegoni, Kufareva (CR9) 2015; 55 de Luca, Horvath, Marcou (CR24) 2012; 52 Vapnik (CR29) 2000 CR28 CR26 CR25 CR23 Iqbal, Vogt, Bajorath (CR10) 2021; 1 Chen, Vogt, Bajorath (CR13) 2022 Pedregosa, Varoquaux, Gramfort (CR27) 2011; 12 Wawer, Bajorath (CR19) 2011; 54 Kingma, Ba (CR34) 2014 Mendez, Gaulton, Bento (CR16) 2019; 47 Ralaivola, Swamidass, Saigo, Baldi (CR30) 2005; 18 Maggiora (CR1) 2006; 46 Tamura, Miyao, Funatsu (CR6) 2020; 39 H Hu (676_CR21) 2019; 11 J Park (676_CR12) 2022 I Maeda (676_CR35) 2022; 36 676_CR15 676_CR37 H Chen (676_CR13) 2022 VN Vapnik (676_CR29) 2000 676_CR33 D Stumpfe (676_CR20) 2012; 55 GM Maggiora (676_CR1) 2006; 46 D Stumpfe (676_CR2) 2014; 57 D Horvath (676_CR8) 2016; 56 J Jiménez-Luna (676_CR14) 2022; 62 J Hussain (676_CR18) 2010; 50 D Stumpfe (676_CR3) 2019; 4 M Wawer (676_CR19) 2011; 54 JH Friedman (676_CR32) 2001 BW Matthews (676_CR38) 1975; 405 F Pedregosa (676_CR27) 2011; 12 J Iqbal (676_CR10) 2021; 1 S Tamura (676_CR6) 2020; 39 K Heikamp (676_CR5) 2012; 52 D Mendez (676_CR16) 2019; 47 J Husby (676_CR9) 2015; 55 676_CR28 676_CR25 J Iqbal (676_CR11) 2021 676_CR26 L Breiman (676_CR31) 2001; 45 676_CR23 L Ralaivola (676_CR30) 2005; 18 S Tamura (676_CR7) 2021; 26 A de Luca (676_CR24) 2012; 52 X Hu (676_CR17) 2012; 52 R Guha (676_CR4) 2012; 52 DP Kingma (676_CR34) 2014 D Rogers (676_CR22) 2010; 50 B Tang (676_CR36) 2020; 12 |
| References_xml | – volume: 12 start-page: 15 year: 2020 ident: CR36 article-title: A self-attention based message passing neural network for predicting molecular lipophilicity and aqueous solubility publication-title: J Cheminformatics doi: 10.1186/s13321-020-0414-z – volume: 47 start-page: D930 year: 2019 end-page: D940 ident: CR16 article-title: ChEMBL: towards direct deposition of bioassay data publication-title: Nucleic Acids Res doi: 10.1093/nar/gky1075 – volume: 36 start-page: 237 year: 2022 end-page: 252 ident: CR35 article-title: Ligand-based approaches to activity prediction for the early stage of structure–activity–relationship progression publication-title: J Comput Aid Mol Des doi: 10.1007/s10822-022-00449-2 – year: 2001 ident: CR32 article-title: Greedy function approximation: a gradient boosting machine publication-title: Ann Stat doi: 10.1214/aos/1013203451 – volume: 50 start-page: 339 year: 2010 end-page: 348 ident: CR18 article-title: Computationally efficient algorithm to identify matched molecular pairs (MMPs) in large data sets publication-title: J Chem Inf Model doi: 10.1021/ci900450m – volume: 52 start-page: 1138 year: 2012 end-page: 1145 ident: CR17 article-title: MMP-cliffs: systematic identification of activity cliffs on the basis of matched molecular pairs publication-title: J Chem Inf Model doi: 10.1021/ci3001138 – volume: 405 start-page: 442 year: 1975 end-page: 451 ident: CR38 article-title: Comparison of the predicted and observed secondary structure of T4 phage lysozyme publication-title: Biochim Biophys Acta doi: 10.1016/0005-2795(75)90109-9 – volume: 26 start-page: 4916 year: 2021 ident: CR7 article-title: Interpretation of ligand-based activity cliff prediction models using the matched molecular pair kernel publication-title: Molecules doi: 10.3390/molecules26164916 – volume: 1 start-page: 100022 year: 2021 ident: CR10 article-title: Learning functional group chemistry from molecular images leads to accurate prediction of activity cliffs publication-title: Artif Intell Life Sci doi: 10.1016/j.ailsci.2021.100022 – volume: 55 start-page: 2932 year: 2012 end-page: 2942 ident: CR20 article-title: Exploring activity cliffs in medicinal chemistry publication-title: J Med Chem doi: 10.1021/jm201706b – ident: CR37 – volume: 62 start-page: 274 year: 2022 end-page: 283 ident: CR14 article-title: Benchmarking molecular feature attribution methods with activity cliffs publication-title: J Chem Inf Model doi: 10.1021/acs.jcim.1c01163 – ident: CR33 – volume: 56 start-page: 1631 year: 2016 end-page: 1640 ident: CR8 article-title: Prediction of activity cliffs using condensed graphs of reaction representations, descriptor recombination, support vector machine classification, and support vector regression publication-title: J Chem Inf Model doi: 10.1021/acs.jcim.6b00359 – year: 2000 ident: CR29 publication-title: The nature of statistical learning theory doi: 10.1007/978-1-4757-3264-1 – volume: 52 start-page: 2181 year: 2012 end-page: 2191 ident: CR4 article-title: Exploring uncharted territories: predicting activity cliffs in structure-activity landscapes publication-title: J Chem Inf Model doi: 10.1021/ci300047k – volume: 12 start-page: 2825 year: 2011 end-page: 2830 ident: CR27 article-title: Scikit-learn: machine learning in python publication-title: J Mach Learn Res – volume: 39 start-page: 2000103 year: 2020 ident: CR6 article-title: Ligand-based activity cliff prediction models with applicability domain publication-title: Mol Inform doi: 10.1002/minf.202000103 – ident: CR25 – volume: 46 start-page: 1535 year: 2006 end-page: 1535 ident: CR1 article-title: On outliers and activity CliffsWhy QSAR often disappoints publication-title: J Chem Inf Model doi: 10.1021/ci060117s – ident: CR23 – year: 2022 ident: CR13 article-title: DeepAC-conditional transformer-based chemical language model for the prediction of activity cliffs formed by bioactive compounds publication-title: Digital Discov doi: 10.1039/d2dd00077f – year: 2021 ident: CR11 article-title: Prediction of activity cliffs on the basis of images using convolutional neural networks publication-title: J Comput Aid Mol Des doi: 10.1007/s10822-021-00380-y – volume: 52 start-page: 2325 year: 2012 end-page: 2338 ident: CR24 article-title: Mining chemical reactions using neighborhood behavior and condensed graphs of reactions approaches publication-title: J Chem Inf Model doi: 10.1021/ci300149n – volume: 54 start-page: 2944 year: 2011 end-page: 2951 ident: CR19 article-title: Local structural changes, global data views: graphical substructure−activity relationship trailing publication-title: J Med Chem doi: 10.1021/jm200026b – volume: 4 start-page: 14360 year: 2019 end-page: 14368 ident: CR3 article-title: Evolving concept of activity cliffs publication-title: ACS Omega doi: 10.1021/acsomega.9b02221 – volume: 50 start-page: 742 year: 2010 end-page: 754 ident: CR22 article-title: Extended-connectivity fingerprints publication-title: J Chem Inf Model doi: 10.1021/ci100050t – volume: 52 start-page: 2354 year: 2012 end-page: 2365 ident: CR5 article-title: Prediction of activity cliffs using support vector machines publication-title: J Chem Inf Model doi: 10.1021/ci300306a – volume: 55 start-page: 1062 year: 2015 end-page: 1076 ident: CR9 article-title: Structure-based predictions of activity cliffs publication-title: J Chem Inf Model doi: 10.1021/ci500742b – ident: CR15 – year: 2014 ident: CR34 article-title: Adam: a method for stochastic optimization publication-title: Arxiv doi: 10.48550/arxiv.1412.6980 – year: 2022 ident: CR12 article-title: ACGCN: graph convolutional networks for activity cliff prediction between matched molecular pairs publication-title: J Chem Inf Model doi: 10.1021/acs.jcim.2c00327 – volume: 11 start-page: 379 year: 2019 end-page: 394 ident: CR21 article-title: Second-generation activity cliffs identified on the basis of target set-dependent potency difference criteria publication-title: Future Med Chem doi: 10.4155/fmc-2018-0299 – volume: 18 start-page: 1093 year: 2005 end-page: 1110 ident: CR30 article-title: Graph kernels for chemical informatics publication-title: Neural Netw doi: 10.1016/j.neunet.2005.07.009 – ident: CR28 – volume: 45 start-page: 5 year: 2001 end-page: 32 ident: CR31 article-title: Random forests publication-title: Mach Learn doi: 10.1023/a:1010933404324 – ident: CR26 – volume: 57 start-page: 18 year: 2014 end-page: 28 ident: CR2 article-title: Recent progress in understanding activity cliffs and their utility in medicinal chemistry publication-title: J Med Chem doi: 10.1021/jm401120g – volume: 39 start-page: 2000103 year: 2020 ident: 676_CR6 publication-title: Mol Inform doi: 10.1002/minf.202000103 – volume: 50 start-page: 339 year: 2010 ident: 676_CR18 publication-title: J Chem Inf Model doi: 10.1021/ci900450m – ident: 676_CR23 – volume: 56 start-page: 1631 year: 2016 ident: 676_CR8 publication-title: J Chem Inf Model doi: 10.1021/acs.jcim.6b00359 – ident: 676_CR25 – volume: 11 start-page: 379 year: 2019 ident: 676_CR21 publication-title: Future Med Chem doi: 10.4155/fmc-2018-0299 – volume: 4 start-page: 14360 year: 2019 ident: 676_CR3 publication-title: ACS Omega doi: 10.1021/acsomega.9b02221 – volume: 47 start-page: D930 year: 2019 ident: 676_CR16 publication-title: Nucleic Acids Res doi: 10.1093/nar/gky1075 – year: 2021 ident: 676_CR11 publication-title: J Comput Aid Mol Des doi: 10.1007/s10822-021-00380-y – volume: 50 start-page: 742 year: 2010 ident: 676_CR22 publication-title: J Chem Inf Model doi: 10.1021/ci100050t – volume: 52 start-page: 2181 year: 2012 ident: 676_CR4 publication-title: J Chem Inf Model doi: 10.1021/ci300047k – year: 2022 ident: 676_CR13 publication-title: Digital Discov doi: 10.1039/d2dd00077f – volume: 52 start-page: 1138 year: 2012 ident: 676_CR17 publication-title: J Chem Inf Model doi: 10.1021/ci3001138 – year: 2014 ident: 676_CR34 publication-title: Arxiv doi: 10.48550/arxiv.1412.6980 – volume: 55 start-page: 1062 year: 2015 ident: 676_CR9 publication-title: J Chem Inf Model doi: 10.1021/ci500742b – volume: 52 start-page: 2354 year: 2012 ident: 676_CR5 publication-title: J Chem Inf Model doi: 10.1021/ci300306a – volume: 18 start-page: 1093 year: 2005 ident: 676_CR30 publication-title: Neural Netw doi: 10.1016/j.neunet.2005.07.009 – volume-title: The nature of statistical learning theory year: 2000 ident: 676_CR29 doi: 10.1007/978-1-4757-3264-1 – volume: 36 start-page: 237 year: 2022 ident: 676_CR35 publication-title: J Comput Aid Mol Des doi: 10.1007/s10822-022-00449-2 – ident: 676_CR28 doi: 10.1145/3292500.3330701 – year: 2001 ident: 676_CR32 publication-title: Ann Stat doi: 10.1214/aos/1013203451 – volume: 1 start-page: 100022 year: 2021 ident: 676_CR10 publication-title: Artif Intell Life Sci doi: 10.1016/j.ailsci.2021.100022 – volume: 62 start-page: 274 year: 2022 ident: 676_CR14 publication-title: J Chem Inf Model doi: 10.1021/acs.jcim.1c01163 – volume: 52 start-page: 2325 year: 2012 ident: 676_CR24 publication-title: J Chem Inf Model doi: 10.1021/ci300149n – volume: 12 start-page: 15 year: 2020 ident: 676_CR36 publication-title: J Cheminformatics doi: 10.1186/s13321-020-0414-z – ident: 676_CR26 – volume: 46 start-page: 1535 year: 2006 ident: 676_CR1 publication-title: J Chem Inf Model doi: 10.1021/ci060117s – volume: 45 start-page: 5 year: 2001 ident: 676_CR31 publication-title: Mach Learn doi: 10.1023/a:1010933404324 – ident: 676_CR37 doi: 10.1109/ICPR.2010.764 – volume: 54 start-page: 2944 year: 2011 ident: 676_CR19 publication-title: J Med Chem doi: 10.1021/jm200026b – year: 2022 ident: 676_CR12 publication-title: J Chem Inf Model doi: 10.1021/acs.jcim.2c00327 – ident: 676_CR33 – volume: 26 start-page: 4916 year: 2021 ident: 676_CR7 publication-title: Molecules doi: 10.3390/molecules26164916 – volume: 57 start-page: 18 year: 2014 ident: 676_CR2 publication-title: J Med Chem doi: 10.1021/jm401120g – ident: 676_CR15 doi: 10.26434/chemrxiv-2022-mfq52-v3 – volume: 55 start-page: 2932 year: 2012 ident: 676_CR20 publication-title: J Med Chem doi: 10.1021/jm201706b – volume: 405 start-page: 442 year: 1975 ident: 676_CR38 publication-title: Biochim Biophys Acta doi: 10.1016/0005-2795(75)90109-9 – volume: 12 start-page: 2825 year: 2011 ident: 676_CR27 publication-title: J Mach Learn Res |
| SSID | ssj0065707 |
| Score | 2.4054763 |
| Snippet | Activity cliffs (AC) are formed by pairs of structural analogues that are active against the same target but have a large difference in potency. While much of... Abstract Activity cliffs (AC) are formed by pairs of structural analogues that are active against the same target but have a large difference in potency. While... |
| SourceID | doaj pubmedcentral proquest gale pubmed crossref springer |
| SourceType | Open Website Open Access Repository Aggregation Database Index Database Enrichment Source Publisher |
| StartPage | 4 |
| SubjectTerms | Accuracy Activity cliff Artificial neural networks Chemistry Chemistry and Materials Science Classifiers Cliffs Comparative analysis Complexity Compound pair-based prediction Computational Biology/Bioinformatics Computer Applications in Chemistry Decision trees Deep learning Documentation and Information in Chemistry Kernel functions Large-scale analysis Learning algorithms Machine learning Memory Methods Neural networks Predictions Support vector machines Theoretical and Computational Chemistry |
| SummonAdditionalLinks | – databaseName: DOAJ - Directory of Open Access Journals dbid: DOA link: http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV3db9QwDI_QhAQvaHyXDRQQEg9QLdePOHkcExNI04T4kPYWpfkYh0bvdL1D_PnYaXusQ8ALr7VTNbbj2I3zM2PPnRXSEhSiqAByzL9c3kDlc2GFjqCcsDbhzJ7A6ak6O9PvL7X6opqwHh64F9xBiLIAF8pCxVApB1pq8FbK0DTOe9-Q9xWgx2Sq98FUzwHjFRklDzrMxApMmzHxIvcsc5hsQwmt_3effGlTuloweeXUNG1Gx7vs1hBF8sP-62-za6G9w24cjc3b7rKvJ1ThnXeogcCXKzqNIQ3wReR0k4EaRnB3MY-x41T5fs6_paLKwG3ruQ9hyYd2EkhJPaY7GjpvKcZM_KkUPfzA99xjn4_ffDp6mw9tFXIH9WydBw--bJzS3oLTTUnXKIsYcSE2upShtLW2OjqhfWFx9UcoHIYpoVReBFEXrrzPdtpFGx4yXvrKS187FUtZ-aqxsnYYcdYQfRMD6IzNRikbN2COU-uLC5NyDyVNrxmDmjFJMwYy9nI7ZtkjbvyV-zUpb8tJaNnpAdqQGWzI_MuGMvaMVG8ID6Olgptzu-k68-7jB3MIJYG4QYVzeTEwxQXOwdnh_gJKgiC0Jpz7E05UvZuSRwszg8PoTAFyJiXGWkh-uiXTSCqCa8Ni0_NohcZeZ-xBb5DbeZcYZ80KUWUMJqY6EcyU0s6_JDhxrQjGTmTs1WjUvz7rz4J_9D8Ev8duFhg0pl9asM921qtNeMyuu-_rebd6kpb0T74vUTk priority: 102 providerName: Directory of Open Access Journals – databaseName: Springer Nature OA Free Journals dbid: C24 link: http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV3rj9MwDI_gQIIvvB-FAwWExAeoyJo2Tj4eEyeQTifEQ7pvUZrHGDraad0Qfz5x2g56PCT4OjvTYjuOvdg_E_LEGiYMQiGyEiCP-ZfNayhdzgxTAaRlxiSc2SM4PpYnJ-rt0BTWjdXu45Nk8tTpWEvxoovZVBFT35g8oYsVOZwnFxBODAu55tjj0PtfrOWAsT3mt-smV1BC6v_VH_90IZ0tljzzYpouosOr_7eFa-TKEHjSg95SrpNzvrlBLs3HeW83yecjLArPu6g0T1drfMBBpdE2UGx-wBkT1J4uQ-goFssv6JdUh-mpaRx13q_oMIEiUtJY6g6XLhsMSxN_ql733-L33CIfD199mL_Oh0kMuYVqtsm9A8drK5UzYFXNsfOyCCGe3Vpx4bmplFHBMuUKEx1GgMLGyMZz6ZhnVWH5bbLXtI2_Syh3pROusjJwUbqyNqKyMUitILg6eFAZmY3K0XaAKcdpGac6pStS6F6KOkpRJylqyMiz3ZpVD9LxV-6XqPMdJwJspw_a9UIP51X7IAqwnhcy-FJaUEKBM0L4urbOuTojj9FiNEJoNFijszDbrtNv3r_TB8AR9w3KuJenA1No4x6sGVoeoiQQdWvCuT_hjKq3U_JomHrwMZ0uQMyEiOFZJD_akXEl1s01vt32PEoyUFVG7vR2vNs3j6HZrGBlRmBi4RPBTCnN8lNCIFcSke9YRp6Pdv7jZ_1Z8Pf-jf0-uVzEiDL93wX7ZG-z3voH5KL9ull264fpzH8HPFBSjA priority: 102 providerName: Springer Nature |
| Title | Large-scale prediction of activity cliffs using machine and deep learning methods of increasing complexity |
| URI | https://link.springer.com/article/10.1186/s13321-022-00676-7 https://www.ncbi.nlm.nih.gov/pubmed/36611204 https://www.proquest.com/docview/2761662449 https://www.proquest.com/docview/2761980795 https://pubmed.ncbi.nlm.nih.gov/PMC9825040 https://doaj.org/article/ef627ce328fe48c79697da66ebbcdddb |
| Volume | 15 |
| WOSCitedRecordID | wos000910831500001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVADU databaseName: BioMed Central_OA刊 customDbUrl: eissn: 1758-2946 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0065707 issn: 1758-2946 databaseCode: RBZ dateStart: 20090101 isFulltext: true titleUrlDefault: https://www.biomedcentral.com/search/ providerName: BioMedCentral – providerCode: PRVAON databaseName: DOAJ - Directory of Open Access Journals customDbUrl: eissn: 1758-2946 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0065707 issn: 1758-2946 databaseCode: DOA dateStart: 20090101 isFulltext: true titleUrlDefault: https://www.doaj.org/ providerName: Directory of Open Access Journals – providerCode: PRVHPJ databaseName: ROAD: Directory of Open Access Scholarly Resources customDbUrl: eissn: 1758-2946 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0065707 issn: 1758-2946 databaseCode: M~E dateStart: 20090101 isFulltext: true titleUrlDefault: https://road.issn.org providerName: ISSN International Centre – providerCode: PRVPQU databaseName: Advanced Technologies & Aerospace Database customDbUrl: eissn: 1758-2946 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0065707 issn: 1758-2946 databaseCode: P5Z dateStart: 20090101 isFulltext: true titleUrlDefault: https://search.proquest.com/hightechjournals providerName: ProQuest – providerCode: PRVPQU databaseName: Health Medical collection customDbUrl: eissn: 1758-2946 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0065707 issn: 1758-2946 databaseCode: 7X7 dateStart: 20090101 isFulltext: true titleUrlDefault: https://search.proquest.com/healthcomplete providerName: ProQuest – providerCode: PRVPQU databaseName: ProQuest Central customDbUrl: eissn: 1758-2946 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0065707 issn: 1758-2946 databaseCode: BENPR dateStart: 20090101 isFulltext: true titleUrlDefault: https://www.proquest.com/central providerName: ProQuest – providerCode: PRVAVX databaseName: Springer Nature OA Free Journals customDbUrl: eissn: 1758-2946 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0065707 issn: 1758-2946 databaseCode: C24 dateStart: 20090112 isFulltext: true titleUrlDefault: https://link.springer.com/search?facet-content-type=%22Journal%22 providerName: Springer Nature |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1LbxMxELZoiwQX3o-FEi0IiQMs3ezDjxNqolRUlCgqIJVeLK8fIajshmyCOPHbmXF2U7aIXrhYSjyOMp7xeMYef0PIc61iqhAKMc4YiyD-0lHBMhPFKhaOcR0r5XFmj9h4zE9OxKQ5cKubtMrWJnpDbSqNZ-R7CcTblMJmJN7Mv0dYNQpvV5sSGltkB5HKQM93BqPx5Li1xZjXwdqnMpzu1RCRJRA-QwCGZppGrLMdedT-v23zH5vTxcTJC7enflM6uPm_7NwiNxp3NNxf689tcsWWd8i1YVsF7i75eoSp4lENorThfIHXOijKsHIhPonAyhOhPps5V4eYQj8Nv_nsTBuq0oTG2nnY1KWAHl-susahsxKdVU_vc9rtT_ide-TTwejj8G3U1GeINMv7y8gaZtJCc2EU06JI8T1m4hys6EKk1KYqF0o4HQuTKDAjjiUa_B2bchPbOE90ep9sl1VpH5IwNZmhJtfcpTQzWaForsF1zZkzhbNMBKTfiknqBrwca2icSR_EcCrXopUgWulFK1lAXm7GzNfQHZdSD1D6G0qE3fZfVIupbFaxtI4mTNs04c5mXDNBBTOKUlsU2hhTBOQZ6o5EYI0SM3emalXX8vDDsdxnKaLBsQx4edEQuQp40Kp5CAEzgVhcHcrdDiWIXne7W92SjeWp5bliBeTpphtHYjZdaavVmkbwmIk8IA_WGr3hOwWHrZ_EWUBYR9c7E9PtKWdfPC654IiHFwfkVbsqzv_Wvyf-0eVcPCbXE_Ar_akX2yXby8XKPiFX9Y_lrF70yBY7Yb7lvWbtw6dhkvX84Qq07wave5jQO8H21wjaSX4KtJPD95PPvwGua2fB |
| linkProvider | ProQuest |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMw1V1Jb9QwFLZKQSoX9iVQICAQB4iaOokdHxAqhaqjDqMKitSbcbwMg0oyTGZY_hS_kfecZEqK6K0HruPnaJ7zvS1-CyGPtYqZwlaIccp5BPGXjgqemihWsXA817FSvs_skI9G-eGh2F8hv7paGEyr7HSiV9Sm0viNfINCvM0YGCPxcvo1wqlReLvajdBoYLFnf36HkK1-MXgN7_cJpTtvDrZ3o3aqQKR5tjmPrOEmKXQujOJaFAlWEVLnAIeFSJhNVCaUcDoWhioAv-NUg5W2SW5iG2dUJ_Dcc-Q8uBE09qmC-53mxywS3hXm5GyjhviPQrAO4R4aBRbxnvHzMwL-tgR_mMKTaZon7mq9Cdy5_L8d3hVyqXW2w61GOq6SFVteI2vb3Yy76-TzEBPhoxqAasPpDC-tEKhh5UIs-MC5GqE-mjhXh1ggMA6_-NxTG6rShMbaadhO3YAVP4q7xq2TEl1xT-8z9u0PeM4N8uFMWL1JVsuqtLdJmJjUMJPp3CUsNWmhWKbBMc-4M4WzXARks4OF1G1rdpwQciR9iJYz2UBJApSkh5LkAXm23DNtGpOcSv0K0bakxKbi_odqNpatjpLWMcq1TWjubJprLpjgRjFmi0IbY4qAPEKsSmwbUmJe0lgt6loO3r-TWzzBXnc8BV6etkSuAh60ass84CSw01iPcr1HCa9e95c7LMtWr9byGMgBebhcxp2YK1jaatHQiDzmIgvIrUaClnwn4I6C3KYB4T3Z6h1Mf6WcfPJd10WO3f7igDzvpPD4b_374O-czsUDsrZ78HYoh4PR3l1ykYIH7b_v8XWyOp8t7D1yQX-bT-rZfa9pQvLxrKXzN5qwumg |
| linkToPdf | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMw1V1Jb9NAFB6VgoAL-2IoYBCIA1hxxvYsB4RKS0TUKIqAShWXYTxLCCp2iBOWv8avY97YTnERvfXANfPGyht_b_O8BaHHSsZEQivEOKU0cvGXinKa6iiWMbeUqVhK32d2RMdjdnDAJxvoV1sLA2mVrU70ilqXCr6R97CLtwlxxoj3bJMWMdkdvJx_jWCCFNy0tuM0aojsmZ_fXfhWvRjuunf9BOPB6_c7b6JmwkCkaNZfRkZTneSKcS2p4nkCFYXYWofJnCfEJDLjklsVc42lEwRLsXIW2yRMxybOsErcc8-gszQlDNLJJtmH1gpARglti3QY6VUuFsQucHehHxgIEtGOIfTzAv62Cn-YxeMpm8fubb05HFz-nw_yCrrUOOHhdi01V9GGKa6hCzvt7Lvr6PMIEuSjygHYhPMFXGYBgMPShlAIAvM2QnU4s7YKoXBgGn7xOakmlIUOtTHzsJnG4Vb8iO4Kts4KcNE9vc_kNz_cc26g_VNh9SbaLMrC3EZholNNdKaYTUiq01ySTDmHPaNW59ZQHqB-CxGhmpbtMDnkUPjQjRFRw0o4WAkPK0ED9Gy9Z143LDmR-hUgb00Jzcb9D-ViKhrdJYwlmCqTYGZNyhTlhFMtCTF5rrTWeYAeAW4FtBMpAE9TuaoqMXz3VmzTBHrg0dTx8rQhsqXjQcmm_MOdBHQg61BudSjdq1fd5RbXotG3lTgCdYAerpdhJ-QQFqZc1TScxZRnAbpVS9Oa78S5qX0cpwGiHTnrHEx3pZh98t3YOYMugHGAnrcSefS3_n3wd07m4gE674RSjIbjvbvoInaOtf_sR7fQ5nKxMvfQOfVtOasW973SCdHH0xbO39EXw1I |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Large-scale+prediction+of+activity+cliffs+using+machine+and+deep+learning+methods+of+increasing+complexity&rft.jtitle=Journal+of+cheminformatics&rft.au=Tamura%2C+Shunsuke&rft.au=Miyao%2C+Tomoyuki&rft.au=Bajorath%2C+J%C3%BCrgen&rft.date=2023-01-07&rft.pub=Springer+Nature+B.V&rft.eissn=1758-2946&rft.volume=15&rft.issue=1&rft.spage=4&rft_id=info:doi/10.1186%2Fs13321-022-00676-7&rft.externalDBID=HAS_PDF_LINK |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1758-2946&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1758-2946&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1758-2946&client=summon |