Back to Search
Start Over
Penalized partial least squares for pleiotropy
- Source :
- BMC Bioinformatics, Vol 22, Iss 1, Pp 1-31 (2021), BMC Bioinformatics, BMC Bioinformatics, 2021, 22 (1), pp.86. ⟨10.1186/s12859-021-03968-1⟩, BMC Bioinformatics, BioMed Central, 2021, 22 (1), pp.86. ⟨10.1186/s12859-021-03968-1⟩
- Publication Year :
- 2021
- Publisher :
- BMC, 2021.
-
Abstract
- Background The increasing number of genome-wide association studies (GWAS) has revealed several loci that are associated to multiple distinct phenotypes, suggesting the existence of pleiotropic effects. Highlighting these cross-phenotype genetic associations could help to identify and understand common biological mechanisms underlying some diseases. Common approaches test the association between genetic variants and multiple traits at the SNP level. In this paper, we propose a novel gene- and a pathway-level approach in the case where several independent GWAS on independent traits are available. The method is based on a generalization of the sparse group Partial Least Squares (sgPLS) to take into account groups of variables, and a Lasso penalization that links all independent data sets. This method, called joint-sgPLS, is able to convincingly detect signal at the variable level and at the group level. Results Our method has the advantage to propose a global readable model while coping with the architecture of data. It can outperform traditional methods and provides a wider insight in terms of a priori information. We compared the performance of the proposed method to other benchmark methods on simulated data and gave an example of application on real data with the aim to highlight common susceptibility variants to breast and thyroid cancers. Conclusion The joint-sgPLS shows interesting properties for detecting a signal. As an extension of the PLS, the method is suited for data with a large number of variables. The choice of Lasso penalization copes with architectures of groups of variables and observations sets. Furthermore, although the method has been applied to a genetic study, its formulation is adapted to any data with high number of variables and an exposed a priori architecture in other application fields.
- Subjects :
- Pathway analysis
Generalization
Computer science
High dimensional data
[SDV]Life Sciences [q-bio]
Genome-wide association study
computer.software_genre
01 natural sciences
Biochemistry
[INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
010104 statistics & probability
Lasso (statistics)
[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing
Structural Biology
Partial least squares regression
Sparse methods
Genetic epidemiology
lcsh:QH301-705.5
0303 health sciences
Lasso Penalization
Applied Mathematics
Methodology Article
Metaanalysis
Computer Science Applications
[SDV] Life Sciences [q-bio]
Variable (computer science)
Phenotype
Oncology
Benchmark (computing)
lcsh:R858-859.7
data processing
Clustering high-dimensional data
Variable selection
sparse coding
Correlation and dependence
Feature selection
Machine learning
lcsh:Computer applications to medicine. Medical informatics
Polymorphism, Single Nucleotide
03 medical and health sciences
statistical analysis
0101 mathematics
Least-Squares Analysis
Partial Least Square
Molecular Biology
030304 developmental biology
Pleiotropy
business.industry
[INFO.INFO-NA]Computer Science [cs]/Numerical Analysis [cs.NA]
Meta-analysis
lcsh:Biology (General)
Artificial intelligence
[INFO.INFO-BI]Computer Science [cs]/Bioinformatics [q-bio.QM]
business
computer
Genome-Wide Association Study
Subjects
Details
- Language :
- English
- ISSN :
- 14712105
- Volume :
- 22
- Issue :
- 1
- Database :
- OpenAIRE
- Journal :
- BMC Bioinformatics
- Accession number :
- edsair.doi.dedup.....f3124ae67ef0ddb1871055e51d0fb70c