Application of the Architecture Transformer for Problem of Recognition of DNA Secondary Structures

Student: Nichvidiuk Daria

Educational Programme: Applied Mathematics and Information Science (Bachelor)

Year of Graduation: 2021

The DNA secondary structure is a form of interaction between the nucleotides of the primary structure of DNA. This paper is a study of the use of various models of the Transformer architecture, trained on natural language corpora, as well as nucleotide chains, to solve the problem of recognizing DNA secondary structure, namely, the lefthanded Z-DNA helix in the genome of the Mus musculus (House mouse) in the mm10 assembly. Applications of the BERT, distilBERT, RoBERTa, XMLRoBERTa, CamemBERT, and ESM-1 models were trained. The paper is aimed at studying and comparing the use of methods that are designed to increase the speed of prediction without losing the accuracy of recognition.

Student Theses at HSE must be completed in accordance with the University Rules and regulations specified by each educational programme.

Summaries of all theses must be published and made freely available on the HSE website.

The full text of a thesis can be published in open access on the HSE website only if the authoring student (copyright holder) agrees, or, if the thesis was written by a team of students, if all the co-authors (copyright holders) agree. After a thesis is published on the HSE website, it obtains the status of an online publication.

Student theses are objects of copyright and their use is subject to limitations in accordance with the Russian Federation’s law on intellectual property.

In the event that a thesis is quoted or otherwise used, reference to the author’s name and the source of quotation is required.

Search all student theses