PhD Scientific Days 2023

Molecular Sciences III.

Network Dynamical Framework to Model Sponging Effect in Post-transcriptional Regulation and to Improve Non-coding RNA Target Prediction

Előadó neve

Benczik, Bettina, MSc

Neptun code

H209EX

Előadó munkahelye

Department of Pharmacology and Pharmacotherapy Semmelweis University, Faculty of Medicine

Előadó telefonszáma

+36309929832

Előadó e-mail címe

benczik.bettina@med.semmelweis-univ.hu

Az előadás címe

Network Dynamical Framework to Model Sponging Effect in Post-transcriptional Regulation and to Improve Non-coding RNA Target Prediction

Szerző(k) neve és munkahelye

Bettina Benczik1,2, Papp Zsolt1,2, Kristóf Körmendi1, Márton Makai1, Mátyás Pétervári1, Péter Ferdinandy1,2, Bence Ágg1,2
1 Cardiometabolic and MTA-SE System Pharmacology Research Group, Department of Pharmacology and Pharmacotherapy, Semmelweis University, Budapest, Hungary
2 Pharmahungary Group, Szeged, Hungary

Bemutatás módja

Szóbeli

Szekció

Molecular Sciences III.

Language of the presentation

Hungarian

Preferred session

Molecular Sciences

Összefoglaló szövege

Introduction: Previously, a conventional RNA logic-based network theoretical microRNA-target prediction software (miRNAtarget™) was developed and validated in multiple studies. However, microRNA-mediated post-transcriptional regulation is affected by many factors. To be able to model the complex regulatory network, we need to include network dynamics to handle sponging effect of the involved components.
Aims: Here we aimed to construct a model of the post-transcriptional regulation based on a network dynamical target prediction framework and to fine-tune and evaluate its predictive performance.
Methods: PRINCE (PRIoritizatioN and Complex Elucidation), a network dynamical algorithm was selected to model the propagation of perturbation induced by non-coding RNA expression changes. PRINCE was applied on non-coding RNA-target interaction networks constructed by the extended miRNAtarget™ to simulate sponging effect and we measured the predictive performance of the novel framework. Quality restrictions were defined to select our validation datasets. Eventually, our validation set consisted of 4 datasets including both small RNA and mRNA sequencing data which overall provided 21 comparisons. Besides the initial miRNAtarget™ software, the performance of two other widely used predictive algorithms (TargetScan, MirTarget) and a random score generation was also compared to our novel approach. For validation, area under the receiver operating characteristic (AUROC) curve analysis was applied followed by DeLong’s test to assign statistical significance to the comparisons of the scores generated by the different algorithms.
Results: Out of the investigated 21 comparisons in 11, PRINCE tendentiously outperformed all the other algorithms out of which in 5 cases it was significant according to DeLong’s test.
Conclusion: Here we demonstrated that our network dynamics-based post-transcriptional regulation model enables the improvement of predictions of mRNA-level expression changes, thus facilitates the identification of novel therapeutic targets.

University and Doctoral School

Semmelweis University, Doctoral School of Pharmaceutical Sciences

Supervisor

Péter Ferdinandy, MD, MBA, PhD; Bence Ágg, MD, PhD

Publication of my abstract

I do not give consent to the publication of my abstract on the website of the congress.

Kind

Szabad

Status

elfogadva

Accepted presentation method

szóbeli

Előadás fájl jóváhagyás

nem rendelkezett róla

Előadó

6896

Start

12:00

End

12:15

Authors (legacy)

Bettina Benczik1,2, Papp Zsolt1,2, Kristóf Körmendi1, Márton Makai1, Mátyás Pétervári1, Péter Ferdinandy1,2, Bence Ágg1,2
1 Cardiometabolic and MTA-SE System Pharmacology Research Group, Department of Pharmacology and Pharmacotherapy, Semmelweis University, Budapest, Hungary
2 Pharmahungary Group, Szeged, Hungary