A novel method for mining highly imbalanced high-throughput screening data in PubChem

Li, Qingliang; Wang, Yanli; Bryant, Stephen H.

doi:10.1093/bioinformatics/btp589

Cited by 56 publications

(65 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Class imbalance occurs frequently in QSAR and drug discovery datasets 14,[65][66][67] . This could be for a number of reasons; however in this context it is due to lack of publically available data for the minority class, poorly-moderately absorbed compounds, in the literature.…”

Section: Resultsmentioning

confidence: 99%

Coping with Unbalanced Class Data Sets in Oral Absorption Models

Newby

Freitas

Ghafourian

2013

J. Chem. Inf. Model.

View full text Add to dashboard Cite

Class imbalance occurs frequently in drug discovery datasets. In oral absorption datasets, in the literature, there are considerably more of highly-absorbed compounds compared with poorly-absorbed compounds. This produces models that are biased towards highly-absorbed compounds which lack generalization to industry settings where more early stage drug candidates are poorly-absorbed. This paper presents two strategies to cope with unbalanced class datasets: Under-sampling the majority high absorption class and misclassification costs using classification decision trees. The published dataset by Hou et al (2007), which contained percentage human intestinal absorption of 645 drug and drug-like compounds, was used for the development and validation of classification trees using C&RT analysis. The results indicate that under-sampling the majority class, highly-absorbed compounds, leads to a balanced distribution (50:50) training set which can achieve better accuracies for poorlyabsorbed compounds, whereas the biased training set achieved higher accuracies for highlyabsorbed compounds. The use of misclassification costs resulted in improved class predictions, when applied to reduce false positives or false negatives. Moreover, it was shown that the classical overall accuracy measure used in many publications is particularly misleading in the case of unbalanced datasets and more appropriate measures presented here may be used for a more realistic assessment of the classification models' performance. Thus, these strategies offer improvements to cope with unbalanced class datasets to obtain classification models applicable in industry.

show abstract

Section: Resultsmentioning

confidence: 99%

Coping with Unbalanced Class Data Sets in Oral Absorption Models

Newby

Freitas

Ghafourian

2013

J. Chem. Inf. Model.

View full text Add to dashboard Cite

show abstract

“…Experiments done on different methods conclude with ambiguous results: while Anand et al [63], certified by Li et al [15] opt for sampling methods as optimal solution, we observe on the other front McCarthy et al [65]in agreement with Liu et al [88] on the superiority of the cost sensitive learning; while Quinlan [94] and Thomas [100] are approving ensemble learning methods; on the other hand Cieslak [38] and Marcellin [90] defend the algorithm modification approaches.…”

Section: Discussionmentioning

confidence: 99%

Imbalanced Data Learning Approaches Review

Bekkar¹,

Alitouche²

2013

IJDKP

142

View full text Add to dashboard Cite

show abstract

“…Additionally, one can use the Pubchem Bioassay search interface to query Pubchem database for relevant biological assay data via PUG REST and SOAP interfaces. 47, 49,50,51,52 A simple search box is presented, the result of a successful search is a list of assays with AID identifiers, short descriptions, and the number of compounds tested for each assay. The AID identifier is a clickable hyperlink which leads to the corresponding assay web page at Pubchem where a user can find more complete information about the assay.…”

Section: Framework and Discussionmentioning

confidence: 99%

Development and implementation of (Q)SAR modeling within the CHARMMing web-user interface

et al. 2014

View full text Add to dashboard Cite

Recent availability of large publicly accessible databases of chemical compounds and their biological activities (PubChem, ChEMBL) has inspired us to develop a Web-based tool for SAR and QSAR modeling to add to the services provided by CHARMMing (www.charmming.org). This new module implements some of the most recent advances in modern machine learning algorithms – Random Forest, Support Vector Machine (SVM), Stochastic Gradient Descent, Gradient Tree Boosting etc. A user can import training data from Pubchem Bioassay data collections directly from our interface or upload his or her own SD files which contain structures and activity information to create new models (either categorical or numerical). A user can then track the model generation process and run models on new data to predict activity.

show abstract

A novel method for mining highly imbalanced high-throughput screening data in PubChem

Cited by 56 publications

References 32 publications

Coping with Unbalanced Class Data Sets in Oral Absorption Models

Coping with Unbalanced Class Data Sets in Oral Absorption Models

Imbalanced Data Learning Approaches Review

Development and implementation of (Q)SAR modeling within the CHARMMing web-user interface

Contact Info

Product

Resources

About