Logistic regression trees for initial selection of interesting loci in case-control studies

Nickolov, Radoslav; Milanov, Valentin

doi:10.1186/1753-6561-1-s1-s57

Cited by 2 publications

(6 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All analyses of the simulated data identified locus C. This locus increased RA risk in women only. Although multiple papers in this group applied machine learning algorithms with a specific interest in identification of interaction effects [Glaser et al, 2007;Meng et al, 2007;Nickolov and Milanov, 2007;Nicodemus et al, 2007;Shi et al, 2007;Stassen et al, 2007;Sun et al, 2007], the interaction of locus C with sex was not detected by methods that specifically modeled this interaction [Meng et al, 2007;Shi et al, 2007]. Locus C was in very strong linkage disequilibrium with the DR loci.…”

Section: Gene Identification Versus Risk Predictionmentioning

confidence: 89%

“…An important distinction is whether the study included the true loci in their search [Nickolov and Milanov, 2007;Nicodemus et al, 2007] or excluded the causative loci and used SNPs in LD for gene detection [Meng et al, 2007;Schwarz et al, 2007;Shi et al, 2007]. All analyses of the simulated data identified locus C. This locus increased RA risk in women only.…”

Section: Gene Identification Versus Risk Predictionmentioning

confidence: 98%

“…LOTUS has been used in only one paper from our group [Nickolov and Milanov, 2007]. However, they demonstrated the usefulness of this simple tree algorithm for reducing the number of SNPs in a case-control study by selecting a small number of interesting SNPs from among a large number of candidate loci.…”

Section: Lotus -Logistic Trees With Unbiased Selectionmentioning

confidence: 98%

“…This locus is a quantitative trait locus for IgM, and it is not clear why this indirect variable was so consistently identified. Loci B and D were identified by those studies that included the true loci [Nickolov and Milanov, 2007;Nicodemus et al, 2007]. These two loci had low linkage disequilibrium between the trait loci and SNP markers which may explain why they were not identified when the true loci were not included [Meng et al, 2007;Schwarz et al, 2007;Shi et al, 2007].…”

Section: S57 Data Mining -Problems 2 and 3 Of Gaw15mentioning

confidence: 98%

“…This seems to have been a general conclusion from the group, because, in a similar vein, both Meng et al [2007] and Schwarz et al [2007] in the simulated and Glaser et al [2007] in the RA data found RF a helpful tool to reduce the number of interesting SNPs as a first step towards gene identification. As an alternative, the same purpose was served well using LOTUS [Nickolov and Milanov, 2007] or the LASSO patternsearch algorithm [Shi et al, 2007] in the simulated data.…”

Section: Gene Identification Versus Risk Predictionmentioning

confidence: 99%

See 4 more Smart Citations

Data mining, neural nets, trees — Problems 2 and 3 of Genetic Analysis Workshop 15

et al. 2007

View full text Add to dashboard Cite

Genome-wide association studies using thousands to hundreds of thousands of single nucleotide polymorphism (SNP) markers and region-wide association studies using a dense panel of SNPs are already in use to identify disease susceptibility genes and to predict disease risk in individuals. Because these tasks become increasingly important, three different data sets were provided for the Genetic Analysis Workshop 15, thus allowing examination of various novel and existing data mining methods for both classification and identification of disease susceptibility genes, gene by gene or gene by environment interaction. The approach most often applied in this presentation group was random forests because of its simplicity, elegance, and robustness. It was used for prediction and for screening for interesting SNPs in a first step. The logistic tree with unbiased selection approach appeared to be an interesting alternative to efficiently select interesting SNPs. Machine learning, specifically ensemble methods, might be useful as pre-screening tools for large-scale association studies because they can be less prone to overfitting, can be less computer processor time intensive, can easily include pair-wise and higher-order interactions compared with standard statistical approaches and can also have a high capability for classification. However, improved implementations that are able to deal with hundreds of thousands of SNPs at a time are required.

show abstract

Section: Gene Identification Versus Risk Predictionmentioning

confidence: 89%

Section: Gene Identification Versus Risk Predictionmentioning

confidence: 98%

Section: Lotus -Logistic Trees With Unbiased Selectionmentioning

confidence: 98%

Section: S57 Data Mining -Problems 2 and 3 Of Gaw15mentioning

confidence: 98%

Section: Gene Identification Versus Risk Predictionmentioning

confidence: 99%

See 3 more Smart Citations

Data mining, neural nets, trees — Problems 2 and 3 of Genetic Analysis Workshop 15

et al. 2007

View full text Add to dashboard Cite

show abstract

RNA-seq assistant: machine learning based methods to identify more transcriptional regulated genes

Wang

Sung

et al. 2018

BMC Genomics

View full text Add to dashboard Cite

BackgroundAlthough different quality controls have been applied at different stages of the sample preparation and data analysis to ensure both reproducibility and reliability of RNA-seq results, there are still limitations and bias on the detectability for certain differentially expressed genes (DEGs). Whether the transcriptional dynamics of a gene can be captured accurately depends on experimental design/operation and the following data analysis processes. The workflow of subsequent data processing, such as reads alignment, transcript quantification, normalization, and statistical methods for ultimate identification of DEGs can influence the accuracy and sensitivity of DEGs analysis, producing a certain number of false-positivity or false-negativity. Machine learning (ML) is a multidisciplinary field that employs computer science, artificial intelligence, computational statistics and information theory to construct algorithms that can learn from existing data sets and to make predictions on new data set. ML–based differential network analysis has been applied to predict stress-responsive genes through learning the patterns of 32 expression characteristics of known stress-related genes. In addition, the epigenetic regulation plays critical roles in gene expression, therefore, DNA and histone methylation data has been shown to be powerful for ML-based model for prediction of gene expression in many systems, including lung cancer cells. Therefore, it is promising that ML-based methods could help to identify the DEGs that are not identified by traditional RNA-seq method.ResultsWe identified the top 23 most informative features through assessing the performance of three different feature selection algorithms combined with five different classification methods on training and testing data sets. By comprehensive comparison, we found that the model based on InfoGain feature selection and Logistic Regression classification is powerful for DEGs prediction. Moreover, the power and performance of ML-based prediction was validated by the prediction on ethylene regulated gene expression and the following qRT-PCR.ConclusionsOur study shows that the combination of ML-based method with RNA-seq greatly improves the sensitivity of DEGs identification.Electronic supplementary materialThe online version of this article (10.1186/s12864-018-4932-2) contains supplementary material, which is available to authorized users.

show abstract

Logistic regression trees for initial selection of interesting loci in case-control studies

Cited by 2 publications

References 11 publications

Data mining, neural nets, trees — Problems 2 and 3 of Genetic Analysis Workshop 15

Data mining, neural nets, trees — Problems 2 and 3 of Genetic Analysis Workshop 15

RNA-seq assistant: machine learning based methods to identify more transcriptional regulated genes

Contact Info

Product

Resources

About