sigFeature

"sigFeature" is a feature selection algorithm designed for classifying biological data into distinct groups, particularly for identifying differentially expressed genes in response to specific treatments. It addresses limitations in existing methods, such as SVM-RFE, by introducing a novel approach that prioritizes biologically significant combinations rather than relying on greedy binary classifications. The algorithm utilizes support vector machines (SVM) to enhance feature selection results' interpretability and biological relevance.

Topic

RNA-Seq;Microarray experiment;Gene expression;Machine learning;Oncology

Detail

  • Operation: Gene-set enrichment analysis;Feature selection;Gene prediction

  • Software interface: Command-line user interface

  • Language: R

  • License: The GNU General Public License v3.0

  • Cost: Free

  • Version name: 1.20.0

  • Credit: The Council of Scientific & Industrial Research.

  • Input: -

  • Output: -

  • Contact: Pijush Das Developer topijush@gmail.com

  • Collection: -

  • Maturity: Stable

Publications

  • sigFeature: Novel Significant Feature Selection Method for Classification of Gene Expression Data Using Support Vector Machine and t Statistic.
  • Das P, et al. sigFeature: Novel Significant Feature Selection Method for Classification of Gene Expression Data Using Support Vector Machine and t Statistic. sigFeature: Novel Significant Feature Selection Method for Classification of Gene Expression Data Using Support Vector Machine and t Statistic. 2020; 11:247. doi: 10.3389/fgene.2020.00247
  • https://doi.org/10.3389/FGENE.2020.00247
  • PMID: 32346383
  • PMC: PMC7169426

Download and documentation

< Back to DB search