PMID- 35936431
OWN - NLM
STAT- PubMed-not-MEDLINE
LR  - 20220809
IS  - 2470-1343 (Electronic)
IS  - 2470-1343 (Linking)
VI  - 7
IP  - 30
DP  - 2022 Aug 2
TI  - Implications of Additivity and Nonadditivity for Machine Learning and Deep 
      Learning Models in Drug Design.
PG  - 26573-26581
LID - 10.1021/acsomega.2c02738 [doi]
AB  - Matched molecular pairs (MMPs) are nowadays a commonly applied concept in drug 
      design. They are used in many computational tools for structure-activity 
      relationship analysis, biological activity prediction, or optimization of 
      physicochemical properties. However, until now it has not been shown in a 
      rigorous way that MMPs, that is, changing only one substituent between two 
      molecules, can be predicted with higher accuracy and precision in contrast to any 
      other chemical compound pair. It is expected that any model should be able to 
      predict such a defined change with high accuracy and reasonable precision. In 
      this study, we examine the predictability of four classical properties relevant 
      for drug design ranging from simple physicochemical parameters (log D and 
      solubility) to more complex cell-based ones (permeability and clearance), using 
      different data sets and machine learning algorithms. Our study confirms that 
      additive data are the easiest to predict, which highlights the importance of 
      recognition of nonadditivity events and the challenging complexity of predicting 
      properties in case of scaffold hopping. Despite deep learning being well suited 
      to model nonlinear events, these methods do not seem to be an exception of this 
      observation. Though they are in general performing better than classical machine 
      learning methods, this leaves the field with a still standing challenge.
CI  - (c) 2022 The Authors. Published by American Chemical Society.
FAU - Kwapien, Karolina
AU  - Kwapien K
AUID- ORCID: 0000-0002-2003-0915
AD  - Medicinal Chemistry, Research and Early Development, Respiratory and Immunology 
      (R&I), BioPharmaceuticals R&D, AstraZeneca, Gothenburg 431 83, Sweden.
FAU - Nittinger, Eva
AU  - Nittinger E
AUID- ORCID: 0000-0001-7231-7996
AD  - Medicinal Chemistry, Research and Early Development, Respiratory and Immunology 
      (R&I), BioPharmaceuticals R&D, AstraZeneca, Gothenburg 431 83, Sweden.
FAU - He, Jiazhen
AU  - He J
AD  - Molecular AI, Discovery Sciences, R&D, AstraZeneca, Gothenburg 431 83, Sweden.
FAU - Margreitter, Christian
AU  - Margreitter C
AUID- ORCID: 0000-0002-5473-6318
AD  - Molecular AI, Discovery Sciences, R&D, AstraZeneca, Gothenburg 431 83, Sweden.
FAU - Voronov, Alexey
AU  - Voronov A
AD  - Molecular AI, Discovery Sciences, R&D, AstraZeneca, Gothenburg 431 83, Sweden.
FAU - Tyrchan, Christian
AU  - Tyrchan C
AUID- ORCID: 0000-0002-6470-984X
AD  - Medicinal Chemistry, Research and Early Development, Respiratory and Immunology 
      (R&I), BioPharmaceuticals R&D, AstraZeneca, Gothenburg 431 83, Sweden.
LA  - eng
PT  - Journal Article
DEP - 20220719
PL  - United States
TA  - ACS Omega
JT  - ACS omega
JID - 101691658
PMC - PMC9352238
COIS- The authors declare no competing financial interest.
EDAT- 2022/08/09 06:00
MHDA- 2022/08/09 06:01
PMCR- 2022/07/19
CRDT- 2022/08/08 03:47
PHST- 2022/05/03 00:00 [received]
PHST- 2022/07/08 00:00 [accepted]
PHST- 2022/08/08 03:47 [entrez]
PHST- 2022/08/09 06:00 [pubmed]
PHST- 2022/08/09 06:01 [medline]
PHST- 2022/07/19 00:00 [pmc-release]
AID - 10.1021/acsomega.2c02738 [doi]
PST - epublish
SO  - ACS Omega. 2022 Jul 19;7(30):26573-26581. doi: 10.1021/acsomega.2c02738. 
      eCollection 2022 Aug 2.