This paper presents the process carried out to obtain the best configuration found for the multiclass support vector machine that allows classifying the mutations present in a set of oligonucleotide microarrays. The proposed solution consists of five phases and we give scope to each phase: obtaining samples, quality control, pre-processing, gene selection and detection of mutations. The samples used were obtained from the biological database GEO-NBCI (Gene Expression Omnibus-National Center for Biotechnology Information), which belong to individuals with the presence of different types of cancer, obtaining 87% accuracy on average for different samples.
|Translated title of the contribution||Detection of oligonucleotide microarray mutations by multiclass support vector machine|
|Number of pages||15|
|Journal||RISTI - Revista Iberica de Sistemas e Tecnologias de Informacao|
|State||Published - Jan 2021|
Bibliographical notePublisher Copyright:
© 2021, Associacao Iberica de Sistemas e Tecnologias de Informacao. All rights reserved.