Machine Learning in Biomedicine

"Uncovering biomarkers via machine learning methods to improve diagnosis, prevention and treatment of diseases in the context of personalize medicine".

DR. MIKEL
RESEARCHER. COMPUTATIONAL BIOLOGY AND TRANSLATIONAL GENOMICS PROGRAM

Home >
Research >
Research divisions >
Research programs >
Computational Biology and Translational Genomics >
Machine Learning in Biomedicine

The Machine learning in biomedicine research group at the Cima Universidad de Navarra currently has two main lines of research:

Analysis of transcriptomic data, both at bulk and single cell resolution.
Development of new file formats for storage and access to omics data.

Molecular biology has undergone a revolution due to the ability to simultaneously study the function and expression of thousands of genes and proteins in the patient's body.

Thanks to the use of computer technology, databases and statistical analysis we can analyze with precision and speed, large amounts of data that allow us to understand the complexity of the mechanisms that cause diseases.

Dr. Mikel Hernáez

GROUP LEADER

	+34 948 194 700
	mhernaez@unav.es

Objectives of the research group in Machine Learning in Biomedicine

Reverse engineering of the transcriptome using machine learning methods

We are currently developing several methods to discover the underlying transcriptional rearrangement, based on both bulk and single-cell RNA-Seq data, and applying them to various diseases.

We anticipate that the methods developed will provide additional new insights into the understanding of key transcriptional rearrangements associated with various cancer diseases, augmented by our ability to follow up in the clinic.

Standards-based compression and encryption of genomic data for cancer research

We are developing a new format for the representation of genomic information that builds on current efforts by the International Organization for Standardization (ISO) to generate a set of specifications for a standards-based representation of these data.

We are also building the software tools that act on the proposed new formats to generate the necessary framework for efficient and secure management of genomic data.

OMICS DATA IN CLINICAL CARE

A challenge for
the not so distant future

Currently, most cancer patients receive treatment minimally informed by omics data; electronic medical records consist of patient-centric data (such as cancer diagnoses, medical history, demographics, imaging, vital signs, lab results, medications, etc.), but cancer genomic data is not stored or accessed through medical records.

In the coming years it will be necessary to store this information in the patient's medical records, so large volumes of data will need to be compressed, stored and accessed securely and efficiently.