Chemical predictive modelling to improve compound quality

Cumming, John G.; Davis, A. M.; Mureşan, Sorel; Haeberlein, Markus; Chen, Hongming

doi:10.1038/nrd4128

Cited by 237 publications

(177 citation statements)

References 105 publications

Supporting

Mentioning

174

Contrasting

Unclassified

Order By: Relevance

“…Many consider the seminal papers of Hansch and Fujita (1964) to be the origin of the QSAR field. Since then, such predictive modelling approaches have grown to become a core part of the drug discovery process (Cumming et al 2013;Cherkasov et al 2014). The subject is still increasing in importance (Cramer 2012).…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Meta-QSAR: a large-scale application of meta-learning to drug design and discovery

et al. 2017

View full text Add to dashboard Cite

We investigate the learning of quantitative structure activity relationships (QSARs) as a case-study of meta-learning. This application area is of the highest societal importance, as it is a key step in the development of new medicines. The standard QSAR learning problem Learn (2018) 107:285-311 is: given a target (usually a protein) and a set of chemical compounds (small molecules) with associated bioactivities (e.g. inhibition of the target), learn a predictive mapping from molecular representation to activity. Although almost every type of machine learning method has been applied to QSAR learning there is no agreed single best way of learning QSARs, and therefore the problem area is well-suited to meta-learning. We first carried out the most comprehensive ever comparison of machine learning methods for QSAR learning: 18 regression methods, 3 molecular representations, applied to more than 2700 QSAR problems. (These results have been made publicly available on OpenML and represent a valuable resource for testing novel meta-learning methods.) We then investigated the utility of algorithm selection for QSAR problems. We found that this meta-learning approach outperformed the best individual QSAR learning method (random forests using a molecular fingerprint representation) by up to 13%, on average. We conclude that meta-learning outperforms base-learning methods for QSAR learning, and as this investigation is one of the most extensive ever comparisons of base and meta-learning methods ever made, it provides evidence for the general effectiveness of meta-learning over base-learning.

show abstract

Section: Discussionmentioning

confidence: 99%

“…A key step in drug development is learning Quantitative Structure Activity Relationships (QSARs) (Martin 2010;Cherkasov et al 2014;Cumming et al 2013). These are functions that predict a compound's bioactivity from its structure.…”

Section: Quantitative Structure Activity Relationship (Qsar) Learningmentioning

confidence: 99%

Meta-QSAR: a large-scale application of meta-learning to drug design and discovery

et al. 2017

View full text Add to dashboard Cite

show abstract

“…Matched Molecular pair (MMP) is defined as a pair of molecules that differ by a minor structural change at a single point [7]. An MMP associated with a significant change in activity is known as "activity cliff" and is of particular interest.…”

Section: Accepted Manuscriptmentioning

confidence: 99%

QSAR models and scaffold-based analysis of non-nucleoside HIV RT inhibitors

Nizami

Tetko

Koorbanally

et al. 2015

Chemometrics and Intelligent Laboratory Systems

View full text Add to dashboard Cite

“…Compound structures were standardized according to AstraZeneca rules 13 and used for compound identification and comparison of compounds tested in both internal and external HTS assays. 4 Compound activity between primary HTS assays screened on the same human-derived protein target (defined by gene symbol) was analyzed using compound activity overlap A 1,2 /(A 1 + A 2 + A 1,2 ), where A 1,2 is the number of compounds active in both assay 1 and assay 2, and A 1 and A 2 are the number of compounds only active in one of the assays and inactive in the other.…”

Section: Compound Activity Analysismentioning

confidence: 99%

Using the BioAssay Ontology for Analyzing High-Throughput Screening Data

et al. 2015

View full text Add to dashboard Cite

High-throughput screening (HTS) is the main starting point for hit identification in drug discovery programs. This has led to a rapid increase of available screening data both within pharmaceutical companies and the public domain. We have used the BioAssay Ontology (BAO) 2.0 for assay annotation within AstraZeneca to enable comparison with external HTS methods. The annotated assays have been analyzed to identify technology gaps, evaluate new methods, verify active hits, and compare compound activity between in-house and PubChem assays. As an example, the binding of a fluorescent ligand to formyl peptide receptor 1 (FPR1, involved in inflammation, for example) in an in-house HTS was measured by fluorescence intensity. In total, 155 active compounds were also tested in an external ligand binding flow cytometry assay, a method not used for in-house HTS detection. Twelve percent of the 155 compounds were found active in both assays. By the annotation of assay protocols using BAO terms, internal and external assays can easily be identified and method comparison facilitated. They can be used to evaluate the effectiveness of different assay methods, design appropriate confirmatory and counterassays, and analyze the activity of compounds for identification of technology artifacts.

show abstract

Chemical predictive modelling to improve compound quality

Cited by 237 publications

References 105 publications

Meta-QSAR: a large-scale application of meta-learning to drug design and discovery

Meta-QSAR: a large-scale application of meta-learning to drug design and discovery

QSAR models and scaffold-based analysis of non-nucleoside HIV RT inhibitors

Using the BioAssay Ontology for Analyzing High-Throughput Screening Data

Contact Info

Product

Resources

About