Deep learning for de novo RNA secondary structure prediction

Marcell Szikszai

doi:10.26182/h67z-s711

Deep learning for de novo RNA secondary structure prediction

Marcell Szikszai

Research output: Thesis › Doctoral Thesis

311 Downloads (Pure)

Abstract

Over the last decade, deep learning has exploded for many applications, ranging from huge advancements in computer vision and natural language processing to biological applications like drug discovery or protein structure prediction (such as DeepMind’s AlphaFold). More recently, deep learning has been applied to RNA secondary structure prediction, with many papers reporting impressive results. This thesis is structured around four main contributions to deep learning for de novo RNA secondary structure prediction: in silico probing experiments, benchmarking existing deep learning models, learning on synthetic data sets, and finally, family prediction and generative models.

Original language	English
Qualification	Doctor of Philosophy
Awarding Institution	The University of Western Australia
Supervisors/Advisors	Ward, Max, Supervisor Wise, Michael, Supervisor Datta, Amitava, Supervisor
Thesis sponsors	Australian Government Research Training Program
Award date	22 Jun 2023
DOIs	https://doi.org/10.26182/h67z-s711
Publication status	Unpublished - 2023

Access to Document

10.26182/h67z-s711

THESIS - DOCTOR OF PHILOSOPHY - SZIKSZAI Marcell - 2023
This work is protected by Copyright. You may print or download ONE copy of this document for the purpose of your own non-commercial research or study. Any other use requires permission from the copyright owner. The Copyright Act requires you to attribute any copyright works you quote or paraphrase.

1 Article

Deep learning models for RNA secondary structure prediction (probably) do not generalise across families
Szikszai, M., Wise, M. J., Datta, A., Ward, M. & Mathews, D., 15 Aug 2022, In: Bioinformatics. 38, 16, p. 3892-3899 8 p.
Research output: Contribution to journal › Article › peer-review

Open Access
26 Citations (Scopus)

Cite this

@phdthesis{c9a63b3365f540e9b187481897298fe7,

title = "Deep learning for de novo RNA secondary structure prediction",

abstract = "Over the last decade, deep learning has exploded for many applications, ranging from huge advancements in computer vision and natural language processing to biological applications like drug discovery or protein structure prediction (such as DeepMind{\textquoteright}s AlphaFold). More recently, deep learning has been applied to RNA secondary structure prediction, with many papers reporting impressive results. This thesis is structured around four main contributions to deep learning for de novo RNA secondary structure prediction: in silico probing experiments, benchmarking existing deep learning models, learning on synthetic data sets, and finally, family prediction and generative models.",

keywords = "machine learning, deep learning, RNA, secondary structure prediction",

author = "Marcell Szikszai",

year = "2023",

doi = "10.26182/h67z-s711",

language = "English",

school = "The University of Western Australia",

}

TY - BOOK

T1 - Deep learning for de novo RNA secondary structure prediction

AU - Szikszai, Marcell

PY - 2023

Y1 - 2023

N2 - Over the last decade, deep learning has exploded for many applications, ranging from huge advancements in computer vision and natural language processing to biological applications like drug discovery or protein structure prediction (such as DeepMind’s AlphaFold). More recently, deep learning has been applied to RNA secondary structure prediction, with many papers reporting impressive results. This thesis is structured around four main contributions to deep learning for de novo RNA secondary structure prediction: in silico probing experiments, benchmarking existing deep learning models, learning on synthetic data sets, and finally, family prediction and generative models.

AB - Over the last decade, deep learning has exploded for many applications, ranging from huge advancements in computer vision and natural language processing to biological applications like drug discovery or protein structure prediction (such as DeepMind’s AlphaFold). More recently, deep learning has been applied to RNA secondary structure prediction, with many papers reporting impressive results. This thesis is structured around four main contributions to deep learning for de novo RNA secondary structure prediction: in silico probing experiments, benchmarking existing deep learning models, learning on synthetic data sets, and finally, family prediction and generative models.

KW - machine learning

KW - deep learning

KW - RNA

KW - secondary structure prediction

U2 - 10.26182/h67z-s711

DO - 10.26182/h67z-s711

M3 - Doctoral Thesis

ER -

Deep learning for de novo RNA secondary structure prediction

Abstract

Access to Document

Fingerprint

Research output

Deep learning models for RNA secondary structure prediction (probably) do not generalise across families

Cite this