Key points are not available for this paper at this time.
We demonstrate that even though PCA ignores the information regarding the class labels of the samples, this unsupervised tool can be remarkably effective as a feature selector. In some cases, it outperforms PLS-DA, which is made aware of the class labels in its input. Our experiments range from looking at the signal-to-noise ratio in the feature selection task, to considering many practical distributions and models encountered when analyzing bioinformatics and clinical data. Other methods were also evaluated. Finally, we analyzed an interesting data set from 396 vaginal microbiome samples where the ground truth for the feature selection was available. All the 3D figures shown in this paper as well as the supplementary ones can be viewed interactively at http://biorg.cs.fiu.edu/plsda CONCLUSIONS: Our results highlighted the strengths and weaknesses of PLS-DA in comparison with PCA for different underlying data models.
Building similarity graph...
Analyzing shared references across papers
Loading...
Daniel Ruiz-Perez
Florida International University
Haibin Guan
Icahn School of Medicine at Mount Sinai
Purnima Madhivanan
University of Arizona
BMC Bioinformatics
SHILAP Revista de lepidopterología
Florida International University
Building similarity graph...
Analyzing shared references across papers
Loading...
Ruiz-Perez et al. (Tue,) studied this question.
synapsesocial.com/papers/69dab05937b5141e3ba3c03c — DOI: https://doi.org/10.1186/s12859-019-3310-7