The problem of predicting RNA secondary structure is a challenging topic, which involves various fields of computer science. Accurate solutions to this problem are helpful in the disciplines of medicine for vaccine development, to design stable mRNA molecules, or biology for discerning between different functions of various RNA molecules according to their shape. The objective of this project is to study an emerging Machine Learning-based approach to the problem of RNA secondary structure prediction via integration of deep learning techniques like transfer learning and convolutional neural net- works, aided by adaptations made for the specific problem at hand, like data representation and loss function. The objective of this project is to provide a new robust Machine Learning-based approach to the problem of RNA secondary structure prediction via integration and improvement of emerging Deep Learning techniques.

The problem of predicting RNA secondary structure is a challenging topic, which involves various fields of computer science. Accurate solutions to this problem are helpful in the disciplines of medicine for vaccine development, to design stable mRNA molecules, or biology for discerning between different functions of various RNA molecules according to their shape. The objective of this project is to study an emerging Machine Learning-based approach to the problem of RNA secondary structure prediction via integration of deep learning techniques like transfer learning and convolutional neural net- works, aided by adaptations made for the specific problem at hand, like data representation and loss function. The objective of this project is to provide a new robust Machine Learning-based approach to the problem of RNA secondary structure prediction via integration and improvement of emerging Deep Learning techniques.

Using transfer learning and loss function adaptation for RNA secondary structure prediction

FALDANI, GIOVANNI
2022/2023

Abstract

The problem of predicting RNA secondary structure is a challenging topic, which involves various fields of computer science. Accurate solutions to this problem are helpful in the disciplines of medicine for vaccine development, to design stable mRNA molecules, or biology for discerning between different functions of various RNA molecules according to their shape. The objective of this project is to study an emerging Machine Learning-based approach to the problem of RNA secondary structure prediction via integration of deep learning techniques like transfer learning and convolutional neural net- works, aided by adaptations made for the specific problem at hand, like data representation and loss function. The objective of this project is to provide a new robust Machine Learning-based approach to the problem of RNA secondary structure prediction via integration and improvement of emerging Deep Learning techniques.
2022
Using transfer learning and loss function adaptation for RNA secondary structure prediction
The problem of predicting RNA secondary structure is a challenging topic, which involves various fields of computer science. Accurate solutions to this problem are helpful in the disciplines of medicine for vaccine development, to design stable mRNA molecules, or biology for discerning between different functions of various RNA molecules according to their shape. The objective of this project is to study an emerging Machine Learning-based approach to the problem of RNA secondary structure prediction via integration of deep learning techniques like transfer learning and convolutional neural net- works, aided by adaptations made for the specific problem at hand, like data representation and loss function. The objective of this project is to provide a new robust Machine Learning-based approach to the problem of RNA secondary structure prediction via integration and improvement of emerging Deep Learning techniques.
Deep learning
Transfer learning
RNA
Loss function
File in questo prodotto:
File Dimensione Formato  
Computer_Engineering_MsC_Thesis___UniPD_GiovanniFaldani.pdf

accesso aperto

Dimensione 3.57 MB
Formato Adobe PDF
3.57 MB Adobe PDF Visualizza/Apri

The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12608/55802