Dataset Information

Complementary PLS and KNN algorithms for improved 3D-QSDAR consensus modeling of AhR binding.

ABSTRACT: Multiple validation techniques (Y-scrambling, complete training/test set randomization, determination of the dependence of R2test on the number of randomization cycles, etc.) aimed to improve the reliability of the modeling process were utilized and their effect on the statistical parameters of the models was evaluated. A consensus partial least squares (PLS)-similarity based k-nearest neighbors (KNN) model utilizing 3D-SDAR (three dimensional spectral data-activity relationship) fingerprint descriptors for prediction of the log(1/EC50) values of a dataset of 94 aryl hydrocarbon receptor binders was developed. This consensus model was constructed from a PLS model utilizing 10 ppm x 10 ppm x 0.5 Å bins and 7 latent variables (R2test of 0.617), and a KNN model using 2 ppm x 2 ppm x 0.5 Å bins and 6 neighbors (R2test of 0.622). Compared to individual models, improvement in predictive performance of approximately 10.5% (R2test of 0.685) was observed. Further experiments indicated that this improvement is likely an outcome of the complementarity of the information contained in 3D-SDAR matrices of different granularity. For similarly sized data sets of Aryl hydrocarbon (AhR) binders the consensus KNN and PLS models compare favorably to earlier reports. The ability of 3D-QSDAR (three dimensional quantitative spectral data-activity relationship) to provide structural interpretation was illustrated by a projection of the most frequently occurring bins on the standard coordinate space, thus allowing identification of structural features related to toxicity.

SUBMITTER: Slavov SH

PROVIDER: S-EPMC3843526 | biostudies-literature | 2013 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Complementary PLS and KNN algorithms for improved 3D-QSDAR consensus modeling of AhR binding.

Slavov Svetoslav H SH Pearce Bruce A BA Buzatu Dan A DA Wilkes Jon G JG Beger Richard D RD

Journal of cheminformatics 20131121 1

Multiple validation techniques (Y-scrambling, complete training/test set randomization, determination of the dependence of R2test on the number of randomization cycles, etc.) aimed to improve the reliability of the modeling process were utilized and their effect on the statistical parameters of the models was evaluated. A consensus partial least squares (PLS)-similarity based k-nearest neighbors (KNN) model utilizing 3D-SDAR (three dimensional spectral data-activity relationship) fingerprint des ...[more]

PMID: 24257141

Similar Datasets

Project description:Simulation of sophisticated biological models requires considerable computational power. These models typically integrate together numerous biological phenomena such as spatially-explicit heterogeneous cells, cell-cell interactions, cell-environment interactions and intracellular gene networks. The recent advent of programming for graphical processing units (GPU) opens up the possibility of developing more integrative, detailed and predictive biological models while at the same time decreasing the computational cost to simulate those models.We construct a 3D model of epidermal development and provide a set of GPU algorithms that executes significantly faster than sequential central processing unit (CPU) code. We provide a parallel implementation of the subcellular element method for individual cells residing in a lattice-free spatial environment. Each cell in our epidermal model includes an internal gene network, which integrates cellular interaction of Notch signaling together with environmental interaction of basement membrane adhesion, to specify cellular state and behaviors such as growth and division. We take a pedagogical approach to describing how modeling methods are efficiently implemented on the GPU including memory layout of data structures and functional decomposition. We discuss various programmatic issues and provide a set of design guidelines for GPU programming that are instructive to avoid common pitfalls as well as to extract performance from the GPU architecture.We demonstrate that GPU algorithms represent a significant technological advance for the simulation of complex biological models. We further demonstrate with our epidermal model that the integration of multiple complex modeling methods for heterogeneous multicellular biological processes is both feasible and computationally tractable using this new technology. We hope that the provided algorithms and source code will be a starting point for modelers to develop their own GPU implementations, and encourage others to implement their modeling methods on the GPU and to make that code available to the wider community.

Project description:Recent genome-scale ChIP-chip studies of transcription factors have shown that a low percentage of experimentally determined binding sites contain the consensus motif for the immunoprecipitated factor. In most cases, differences between in vivo target sites that contain or lack a consensus motif have not been explored. We have previously shown that most sites to which E2F family members are bound in vivo do not contain E2F consensus motifs. The main purpose of this study was to develop an understanding of how E2F binding specificity is achieved in vivo. In particular, we have addressed how E2F family members are recruited to core promoter regions that lack a consensus motif and are excluded from other regions that contain a consensus motif. Using promoter and ENCODE arrays, we have shown that the predominant factors specifying whether E2F is recruited to an in vivo binding site are a) the site must be in a core promoter and b) the promoter region must be utilized as a promoter by the transcriptional machinery in that particular cell type. We have tested three models for recruitment of E2F to core promoters lacking a consensus site, including a) indirect recruitment, b) looping to the core promoter mediated by an E2F bound to a distal consensus motif, and c) assisted binding of E2F to a site that weakly resembles an E2F consensus motif within the core promoter. To test these models, we developed a new in vivo assay, termed eChIP, which allows analysis of transcription factor binding to isolated promoter fragments. Our findings suggest that in vivo a) the presence of a consensus motif is not sufficient to recruit E2Fs, b) E2Fs can bind to isolated regions that lack a consensus motif, and c) binding can require regions other than the best match to the E2F PWM in the core promoter. Keywords: E2F, ChIP-chip, transcription factor binding, consensus motifs 37 ChIP-chip arrays (of these, 14 array sets are biological duplicates). 22 samples are included in this series, the rest can be found in supplementary info to the following papers: Xu 2007, Jin 2006, Komashko 2008

Project description:BackgroundChanges in microRNA (miRNA) expression patterns have been extensively characterized in several cancers, including human colon cancer. However, how these miRNAs and their putative mRNA targets contribute to the etiology of cancer is poorly understood. In this work, a bioinformatics computational approach with miRNA and mRNA expression data was used to identify the putative targets of miRNAs and to construct association networks between miRNAs and mRNAs to gain some insights into the underlined molecular mechanisms of human colon cancer.MethodThe miRNA and mRNA microarray expression profiles from the same tissues including 7 human colon tumor tissues and 4 normal tissues, collected by the Broad Institute, were used to identify significant associations between miRNA and mRNA. We applied the partial least square (PLS) regression method and bootstrap based statistical tests to the joint expression profiles of differentially expressed miRNAs and mRNAs. From this analysis, we predicted putative miRNA targets and association networks between miRNAs and mRNAs. Pathway analysis was employed to identify biological processes related to these miRNAs and their associated predicted mRNA targets.ResultsMost significantly associated up-regulated mRNAs with a down-regulated miRNA identified by the proposed methodology were considered to be the miRNA targets. On average, approximately 16.5% and 11.0% of targets predicted by this approach were also predicted as targets by the common prediction algorithms TargetScan and miRanda, respectively. We demonstrated that our method detects more targets than a simple correlation based association. Integrative mRNA:miRNA predictive networks from our analysis were constructed with the aid of Cytoscape software. Pathway analysis validated the miRNAs through their predicted targets that may be involved in cancer-associated biological networks.ConclusionWe have identified an alternative bioinformatics approach for predicting miRNA targets in human colon cancer and for reverse engineering the miRNA:mRNA network using inversely related mRNA and miRNA joint expression profiles. We demonstrated the superiority of our predictive method compared to the correlation based target prediction algorithm through a simulation study. We anticipate that the unique miRNA targets predicted by the proposed method will advance the understanding of the molecular mechanism of colon cancer and will suggest novel therapeutic targets after further experimental validations.

Project description:Mathematics plays a leading part in day-to-day life and has enhanced a necessary component for human accomplishments. Students from many countries do not reach the expected level in mathematics. Therefore, it is essential to pay close consideration to the causes related to ability in mathematics. Mathematics attitude is considered as one of the critical variables in the process of mathematics learning. This study aimed to determine students' attitudes and achievements through problem-based learning in mathematics. The selected study group contained 600 students and 35 teachers from rural public secondary schools in District Rawalpindi, Pakistan. The data collection was done using questionnaires from students and teachers and collected data analyzed by SPSS 23 and Amos 23. This study's result was carried out using Partial Least square structural equation Model (PLS-SEM), descriptive analysis, and hypotheses testing. The outcomes in this study indicated that the mean fluctuated between 1 to 4.5, 3.71 to 4.20, and Std. Deviation fluctuated between 0.6 to 2.0 and 0.75 to 1.55 in the students and teacher models, respectively. The results of the PLS-SEM students' model show a negative attitude towards mathematics. The teachers' PLS-SEM model showed the Effects of using problem-based learning (PBL) on students' achievements. According to the hypotheses testing, the acceptance of hypotheses by stating that the Confidence in Learning Mathematics Scale (C), Value of Mathematics Scale (V), and Student Mathematics Motivation Scale (M) are significant effects for the Students' Attitude Toward Problem-Based Learning (ATPBL). But the Attitude Toward Enjoyment in Mathematics Scale (AE) was rejected, and it did not significantly affect the ATPBL. As well as, the Problem-solving learning and students' achievement (PLA), Advantages of problem-solving learning (APL) and Difficulties in using problem-solving learning (DPL) have a significant positive effect on the ATPBL. Finally, this study suggested that teachers also adopt new teaching methods corresponding to mathematics, and there is a need to explore particular mathematics skills to enhance students' learning abilities.

Dataset Information

Complementary PLS and KNN algorithms for improved 3D-QSDAR consensus modeling of AhR binding.

Publications

Complementary PLS and KNN algorithms for improved 3D-QSDAR consensus modeling of AhR binding.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets