Laboratory of Electronic and Multimedia Applications (Research Section)

[Katarína Grešová]: Modeling Small RNA Binding Rules 30. 3. 2023


Abstract

Small RNAs bind their targets in a sequence and structure-dependent manner. The rules of binding for each category of small RNAs have been studied for a long time, but have not to date been clearly identified. New technological advancements in the fields of Sequencing have allowed the production of 'chimeric' reads that contain both the small RNA and a part of the binding site sequence. Using these types of datasets we can, for the first time, have an unbiased prediction of where small RNAs bind to their targets.

In this work, we will examine existing approaches for the task of small RNA target prediction and describe their strengths and biases. Based on this knowledge, we will propose a method that will be able to predict targets of small RNAs based on sequence alone. We will use the BERT model that can be pretrained on a large amount of data in the broader domain and fine-tuned on smaller amounts of data from a specific domain.

Presentation

PV174 31.3.2022 AgoBind
PDF ke stažení

záznam semináře Katarína Grešová

Readings

    1. Grigoriev, A. (2021). Transfer RNA and Origins of RNA Interference (https://www.frontiersin.org/articles/10.3389/fmolb.2021.708984/full)
    2. Ji, Y., Zhou, Z., Liu, H., & Davuluri, R. V. (2021). DNABERT: pre-trained Bidirectional Encoder Representations from Transformers model for DNA-language in genome (https://academic.oup.com/bioinformatics/article/37/15/2112/6128680)
    Předchozí
    Následující