02 Research Projects

Computational analysis of splicing regulation
Sofia Pinto, Kristian Vlahoviček

This project analyzes regulators of exon splicing, the segments of genes which code for proteins. In mammals, exons are recognized by specific proteins which bind to short sequences in the exons, called the exon splicing enhancers (ESE). If an ESE is not recognized, that exon can be skipped in the process of translation, which makes the final protein product incomplete and can cause certain genetic disorders. Generally, the prediction of ESEs in the genome is difficult because these sequences are very short, degenerat and can be context dependent. The ESE Analyzer web server being developed predicts ESEs by combining alignment results of paralagous sequences of human DNA, conserved motifs and other exon specific information.

PRO-MINE database
Sofia Pinto, Kristian Vlahoviček

This project is developing the PROtein Mutations In NEurodegeneration (PRO-MINE) database which contains manually curated information on all disease-causing mutations in the TDP-43 gene and is publicly available at http://bioinfo.hr/pro-mine/. The TDP-43 nuclear factor is a protein known for a high number of point mutations, and it has a role numerous cellular processes such as transcription and mRNA stability and transport. There are indications that miss-regulation of TDP-43 can cause some neurodegenerative disease. The PRO-MINE database is a repository and an integrated set of tools for computational analysis of genetic diseases in neurodegenerative diseases.

Computational profiles of human glycan profiles
Sofia Pinto, Kristian Vlahoviček

This project analyzes glycan profiles of human blood plasma. We have a variety of high-throughput experimental methods available today which can encompass and create a lot of data from whole biological systems. In parallel, there is a need for efficient data processing to gain as much useful information as possible from these analyses. An example of such a high-throughput experimental method is HPLC analysis of human blood plasma glycan profiles. This project is developing a system of methods based on machine learning which can search through HPLC data and discover patterns of glycolysation within large groups of the population.

Computational analysis of the relationship between histone modifications and gene expression
Rosa Karlić, Kristian Vlahoviček

This project studies histone modifications, or modifications of proteins which participate in packing of DNA and are closely linked to gene regulation. There is evidence that histone modifications are in fact a mean for coding for certain processes downstream of genes. Quantative models for predictions of gene expression levels through histone modification have shown thus far that gene expression and histone modification have a high correlation and that a relatively small number of histone modifications are necessary to predict gene expression levels, as well as that their relationship is general and valid in many cell types.

Codon usage in metagenomes
Vedran Lucić, Maša Roller, Kristian Vlahoviček

Metagenomics, the study of bacterial communities isolated straight from the environment, is a field which investigates bacterial populations in their natural environments. We identify functional adaptations of whole bacterial communities which enable them to survive in diverse environmental conditions such as animal digestive tracts, where bacteria degrade food into forms which can be harvested by the animal and where changes in bacterial composition can lead to obesity and disease. The method we are developing can not only identify the functions bacteria posses but also predict how much they are used and therefore offer a more accurate description of life in bacterial communities.

Genomic complexity of the simplest Metazoa, the sponge
Maša Roller, Kristian Vlahoviček

Sponges are living fossils, inhabiting the plant for 580 million years, and the closest living species to the first multicellular animal. Therefore, by researching sponges we are also investigating our own evolutionary past. Our research of sponge genomes has shown that, despite the morphological simplicity of sponges, their genetic repertoire is surprisingly complex. This shows that even ancient ancestors of animals already had the potential for more complex life functions. The focus in evolutionary studies is shifting from the increase of genome complexity to adaptations which enable exploitation of the existing complexity.

The relationship between DNA flexibility and transcription
Tina Kokan, Kristian Vlahoviček

This project is connecting functional regions of the DNA molecules with its three-dimensional structure. The DNA molecule is a hereditary life molecule that can be found in all living beings, from bacteria to humans. The structure of the DNA molecule is basically a double helix, but it is not equally flexible everywhere. There are small local deviations in the flexibility of the double helix, which have thus far been connected to many crucial cell biological processes, such as chromosome packaging, recombination and DNA duplication. This research is focused on discovering how these small local polymorphisms relate to the biological process of transcribing information from the DNA molecule at the level of whole genomes of humans and fruit flies.

MADNet, a tool for analysis and visualization of data from high-throughput methods
Petar Glažar, Petar Jager, Kristian Vlahoviček

MADNet is a software tool for analysis and visualization of data from high-throughput methods, publicly available at http://madnet.bioinfo.hr/. To reach a scientific hypothesis from a huge amount of raw data it is necessary to consider many different types of information such as gene expression levels, transcription factor regulation, position in metabolic pathways etc. MADNet integrates all this information and presents it visually in metabolic pathway maps, which facilitates a user’s evaluation of data and reduces the number of necessary further experiments.

Written by