Novel mathematical and statistical methods for Machine Learning with applications in the Next Generation Sequencing technologies in Genetics

HOME | PARTICIPANTS | EVENTS | PUBLICATIONS | PRESENTATIONS | ABOUT

 

The goal of this project is to develop novel advanced mathematical and statistical methods applicable to the next generation sequencing (NGS) technologies and to improve some of the existing ones. The NGS technologies that have emerged since the mid-2000s, allow researchers to produce genetic data that is orders of magnitude larger than the previous methods.

Planned tasks are:
- Development of novel statistical tests and computing the statistical power for RNA sequencing data. This include tests for comparison gene expression levels between two groups, e.g. between patients and healthy controls. We expect:
- To calculate the power for Negative Binomial-based differential expression methods developing novel statistical tests and computing the statistical power for DNA sequencing data
- Development of new distances for the specific sequences of Genetics data and application of the new distances to real or simulated genomic datasets
- Development of new methods of Machine Learning, based on the recent developments in Wavelet Analysis and application of newly developed wavelets of Daubechies type to the study of genomic sequences
- Investigation of the polyharmonic kernels and the properties of the Kernel Learning algorithms and adaptation of the Kernel Learning algorithm to some real data, including the genomic sequences.

The main methodology of this project consists of mathematical and statistical methods, related to Genetics problems. The project team members are qualified in appropriated mathematical and statistical methods that will be used to achieve the project goals. In our study, we will derive novel methods for finding solutions of these problems and will also apply some existing ones.

 

 

HOME | PARTICIPANTS | EVENTS | PUBLICATIONS | PRESENTATIONS | ABOUT