Data mining PubChem using a support vector machine with the Signature molecular descriptor: Classification of factor XIa inhibitors

Weis, Derick C.; Visco, Donald P.; Faulon, Jean-Loup

doi:10.1016/j.jmgm.2008.08.004

Cited by 41 publications

(47 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, there is still a large amount of high-throughput screening data and six crystal structures available. 23 In the following, we will first give an impression of the differences in hit retrieval with geometrically identical models using each program's default settings. We define models to be 'geometrically identical' if the same coordinates are used for feature placement and if they match the same active compounds in each software tool.…”

Section: Introductionmentioning

confidence: 99%

One Concept, Three Implementations of 3D Pharmacophore-Based Virtual Screening: Distinct Coverage of Chemical Search Space

Spitzer

Heiss

Mangold

et al. 2010

J. Chem. Inf. Model.

View full text Add to dashboard Cite

Feature-based pharmacophore modeling is a well-established concept to support early stage drug discovery, where large virtual databases are filtered for potential drug candidates. The concept is implemented in popular molecular modeling software, including Catalyst, Phase, and MOE. With these software tools we performed a comparative virtual screening campaign on HSP90 and FXIa, taken from the 'maximum unbiased validation' data set. Despite the straightforward concept that pharmacophores are based on, we observed an unexpectedly high degree of variation among the hit lists obtained. By harmonizing the pharmacophore feature definitions of the investigated approaches, the exclusion volume sphere settings, and the screening parameters, we have derived a rationale for the observed differences, providing insight on the strengths and weaknesses of these algorithms. Application of more than one of these software tools in parallel will result in a widened coverage of chemical space. This is not only rooted in the dissimilarity of feature definitions but also in different algorithmic search strategies.

show abstract

Section: Introductionmentioning

confidence: 99%

One Concept, Three Implementations of 3D Pharmacophore-Based Virtual Screening: Distinct Coverage of Chemical Search Space

Spitzer

Heiss

Mangold

et al. 2010

J. Chem. Inf. Model.

View full text Add to dashboard Cite

show abstract

“…Thus, HTS data is typically imbalanced in general with a small ratio of active compounds to inactive ones. Although many researchers (Guha and Schurer, 2008; Han et al ., 2008; Weis et al ., 2008) have noticed this problem when using the data in PubChem, to the best of our knowledge, there has been no method reported to tackle this problem effectively. Han et al .…”

Section: Introductionmentioning

confidence: 99%

“…Weis et al . (2008) suggested the assay data be carefully selected from PubChem to attempt to avoid using the imbalanced data in their study.…”

Section: Introductionmentioning

confidence: 99%

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

Wang

Bryant

2009

Bioinformatics

View full text Add to dashboard Cite

Motivation: The comprehensive information of small molecules and their biological activities in PubChem brings great opportunities for academic researchers. However, mining high-throughput screening (HTS) assay data remains a great challenge given the very large data volume and the highly imbalanced nature with only small number of active compounds compared to inactive compounds. Therefore, there is currently a need for better strategies to work with HTS assay data. Moreover, as luciferase-based HTS technology is frequently exploited in the assays deposited in PubChem, constructing a computational model to distinguish and filter out potential interference compounds for these assays is another motivation.Results: We used the granular support vector machines (SVMs) repetitive under sampling method (GSVM-RU) to construct an SVM from luciferase inhibition bioassay data that the imbalance ratio of active/inactive is high (1/377). The best model recognized the active and inactive compounds at the accuracies of 86.60% and 88.89 with a total accuracy of 87.74%, by cross-validation test and blind test. These results demonstrate the robustness of the model in handling the intrinsic imbalance problem in HTS data and it can be used as a virtual screening tool to identify potential interference compounds in luciferase-based HTS experiments. Additionally, this method has also proved computationally efficient by greatly reducing the computational cost and can be easily adopted in the analysis of HTS data for other biological systems.Availability: Data are publicly available in PubChem with AIDs of 773, 1006 and 1379.Contact: ywang@ncbi.nlm.nih.gov; bryant@ncbi.nlm.nih.govSupplementary information: Supplementary data are available at Bioinformatics online.

show abstract

“…Schierz studied the effect of false positive problems of the PubChem bioassay on virtual screening [14]. Moreover, the PubChem bioassay data have been used in various modeling studies: a decision tree approach using the HTS data [15]; a Bayesian approach using the HTS data [16]; a support vector machine (SVM) approach for inhibitor or ligand classifications [17]; and a GPU accelerated SVM approach [18]. In 2010, Xie reviewed the previous applications of bioassay data in PubChem and he also addressed two major challenges in the application of PubChem bioassay data: a biased active/inactive ratio and experimental errors shown as false positives or negatives [19].…”

Section: Introductionmentioning

confidence: 99%

Profiling Animal Toxicants by Automatically Mining Public Bioassay Data: A Big Data Approach for Computational Toxicology

2014

View full text Add to dashboard Cite

In vitro bioassays have been developed and are currently being evaluated as potential alternatives to traditional animal toxicity models. Already, the progress of high throughput screening techniques has resulted in an enormous amount of publicly available bioassay data having been generated for a large collection of compounds. When a compound is tested using a collection of various bioassays, all the testing results can be considered as providing a unique bio-profile for this compound, which records the responses induced when the compound interacts with different cellular systems or biological targets. Profiling compounds of environmental or pharmaceutical interest using useful toxicity bioassay data is a promising method to study complex animal toxicity. In this study, we developed an automatic virtual profiling tool to evaluate potential animal toxicants. First, we automatically acquired all PubChem bioassay data for a set of 4,841 compounds with publicly available rat acute toxicity results. Next, we developed a scoring system to evaluate the relevance between these extracted bioassays and animal acute toxicity. Finally, the top ranked bioassays were selected to profile the compounds of interest. The resulting response profiles proved to be useful to prioritize untested compounds for their animal toxicity potentials and form a potential in vitro toxicity testing panel. The protocol developed in this study could be combined with structure-activity approaches and used to explore additional publicly available bioassay datasets for modeling a broader range of animal toxicities.

show abstract

Data mining PubChem using a support vector machine with the Signature molecular descriptor: Classification of factor XIa inhibitors

Cited by 41 publications

References 49 publications

One Concept, Three Implementations of 3D Pharmacophore-Based Virtual Screening: Distinct Coverage of Chemical Search Space

One Concept, Three Implementations of 3D Pharmacophore-Based Virtual Screening: Distinct Coverage of Chemical Search Space

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

Profiling Animal Toxicants by Automatically Mining Public Bioassay Data: A Big Data Approach for Computational Toxicology

Contact Info

Product

Resources

About