Our topic covers two subprojects both connected to the phases of drug development.
We aim to develop softwares based on network theoretic approaches that are capable to identify mediators and pathways involved in the pathomechanism of various cardiovascular diseases by the analysis of datasets assessed with high throughput molecular biological techniques (e.g. microarray, RNA-seq) of samples gathered from small and large animal models of the studied cardiovascular disorders. Our main focus is the analysis of the transcriptomics datasets, including microRNA fingerprints.
In our second subtopic we are searching for network theoretic algorithms that by the analysis of various pharmacovigilance databases could predict adverse drug reactions or adverse interactions of drugs even during the drug development.
Group leader: Peter Ferdinandy, M.D., D.Sc., Head of Department
Publications in the field:
- Ágg B, Baranyai T, Makkos A, Vető B, Faragó N, Zvara Á, Giricz Z, Veres DV, Csermely P, Arányi T, Puskás LG, Varga ZV, Ferdinandy P. MicroRNA interactome analysis predicts post-transcriptional regulation of ADRB2 and PPP3R1 in the hypercholesterolemic myocardium. Sci Rep. 2018 Jul 4;8(1):10134.
- Varga ZV, Ágg B, Ferdinandy P. miR-125b is a protectomiR: A rising star for acute cardioprotection. J Mol Cell Cardiol. 2018 Feb;115:51-53.
- Schulz R, Ágg B, Ferdinandy P. Survival pathways in cardiac conditioning: individual data vs. meta-analyses. What do we learn? Basic Res Cardiol. 2017 Nov 21;113(1):4.
- Perrino C, Barabási AL, Condorelli G, Davidson SM, De Windt L, Dimmeler S, Engel FB, Hausenloy DJ, Hill JA, Van Laake LW, Lecour S, Leor J, Madonna R, Mayr M, Prunier F, Sluijter JPG, Schulz R, Thum T, Ytrehus K, Ferdinandy P. Epigenomic and transcriptomic approaches in the post-genomic era: path to novel targets for diagnosis and therapy of the ischaemic heart? Position Paper of the European Society of Cardiology Working Group on Cellular Biology of the Heart. Cardiovasc Res. 2017 Jun 1;113(7):725-736.
Achievements of undergraduate researchers:
- Balázs Petrovich – Semmelweis University Students’ Scientific Association (TDK) Conference, Budapest, Hungary, 2018 – 1st prize
- Balázs Petrovich – XXIII. Korányi Frigyes Scientific Forum, Budapest, Hungary, 2018 – 1st prize
Unbiased network theoretic microRNA interactome analysis for the prediction of most probable targets of differentially expressed microRNAs in cardiovascular diseases
MicroRNAs (miRNAs) are short (18-24 nt) RNA sequences that could decrease the expression of various target genes both on the messenger RNA (mRNA) and the protein level. As miRNAs could have multiple hundreds of targets, and also one mRNA could be regulated by dozens of miRNAs the post-transcriptional regulatory network of miRNAs are extremely complex (Figure 1). Therefore to understand and predict the effect of changes in the miRNA expression profiles measured by the so called omics (global, high throughput) methodologies, sophisticated bioinformatics and network theoretic approaches are necessary.
In this project we aim to develop, implement and utilize such network theoretic models to efficiently search for drug target candidates in various cardiovascular diseases.
We also aim to make it possible for researchers without previous bioinformatics experiences to analyze miRNA expression datasets with user friendly tools. For this purpose our team in collaboration with the Pharmahungary Group has developed the miRNAtarget.com web application, which could be used to predict common targets of differentially expressed miRNAs.
Figure 1. An example of miRNA-target interaction network. Downregulated, upregulated miRNAs and mRNAs are indicated in green, red and blue, respectively. Dark blue represents mRNAs with at least 4 target interactions. (Ágg et al, Sci Rep 2018)
Learning opportunities: software development skills in C++ and R languages, bioinformatics methodologies, network theory
Project leader: Bence Ágg, MD
Applying network theory models on adverse drug event datasets in order to identify novel cardiovascular effects of marketed drugs
Unidentified and preventable adverse drug interactions yield a heavy burden on the public health sector. These adverse reactions are implying severe and prolonged potential risks especially in case of cardiovascular agents. There are databases containing more than 10 million adverse event reports but the available analysis tools, which could efficiently handle this data, are still limited.
The aim of this project is to build and characterize networks based on the available adverse drug event datasets in order to better understand the casual relationship between drugs and adverse drug reactions which would allow earlier detection of harmful drug effects.
As adverse drug event reports are considered noisy, we need to use complex data purification methods to allow the extraction of valuable information from multiple thousands of unique reports. Afterwards we apply different type of network topological methods in order reveal the previously hidden drug – adverse drug reaction phenomena.
With our model we intend to contribute to the detection of safety signals and high-risk patient sub-populations.
Learning opportunities: Clinical trial and pharmacovigilance documentation/regulation, Clinical data handling, Software development in C++, Network theory