OMICs prediction for Hordeum vulgare using Random Forest methodology

abstract

English

In recent years, due to the growing quality and decreasing cost of sequencing research has shifted to the post-genomic are, in which its main purpose is to understand the relationship between genotype and phenotype. This fact opens new possibilities and reveals a new blank space of missing in silico tools. Current major players for undestanding of functional relationships can represent tools based on machine learning methods, such as random forest. This appraoch brings many interesting results which cannot be connect only with prediction and created prediction models but also feature selection analysis revealed the new weighting priority for follow-up study using more sophisticated algorithms based on deep learning. In this work, the prediction models through spectrum of omics data was created. Especially, the genomic and metabolomic prediction was created using random forest regressor. The studied data are focused on Hordeum vulgare which has recently been widely used in the cosmetics and pharmaceutical industries due to its antioxidant effects.

PANOMICs analysis, Prediction modelling, Random Forest, Hordeum vulgare

SCHWARZEROVÁ, J.; NEMČEKOVÁ, P.; PIERDIES, I.; NOHEL, M.; CHMELÍK, J.; SEDLÁŘ, K.; WECKWERTH, W.

28. 4. 2022

Masaryk University Press

Brno

978-80-280-0040-0

The Biomania Student Scientific Meeting 2022, Book of abstract

1st

52

52

1