In Silico identification of non-coding RNAs in Halobacterium salinarum NRC-1 model archeon organism

Author(s)

Marcos Abraão de Souza Fonseca

• Document type: Doctoral Thesis
• Press: São Paulo.
• Institution: Universidade de São Paulo (USP). Instituto de Matemática e Estatística (IME/SBI)
• Defense date: 2016-04-25

Examining board members

Ricardo Zorzetto Nicoliello Vencio; Angela Kaysel Cruz; Georgios Joannis Pappas Junior; Aline Maria da Silva; Maria Emília Machado Telles Walter

• Advisor: Ricardo Zorzetto Nicoliello Vencio

Abstract

The gene expression regulation occurs on different cell levels in response to dynamics established between an organism and its environment. In addition to the regulatory elements already known, for instance, transcription factors or post-translation modifications, there is growing interests in the regulatory role played by non-coding RNA molecules (ncRNA) whose functions can be performed on different level of biological information processing. Model organisms allow a convenient way to work on laboratory and different research groups aiming to guide their studies for a mutual and wide understanding of the cellular mechanisms present on these organisms. Although some ncRNAs elements have been found in Halobacterium salinarum model organism we believe that not enough is knowing about these genomic regions. In these context, an in silico analysis for ncRNAs identification and RNA-protein prediction approach were applied to H. salinarum NRC-1. Considering a data integration perspective and some available methodologies, several machine learning models was built and used to designate candidate ncRNAs genome regions. According to achieve results, 42 new ncRNAs could be identified, increasing 82% the total of known ncRNAs in H. salinarum NRC-1. Combing analysis with other available tools, it had been observed that some suggested candidates also was found with different methodologies and thus, it highlights the proposed results. Additionally, we developed and analyzed methods, also machine learning based, to predict ncRNAs candidates to interact with LSm protein, present on the interested model organism aiming a basic ncRNA characterization. The achieved results in this part was not satisfactory since the applied models were not substantially accurate predictions. However, we believe that these preliminary results can contribute with some discussions to new different approaches. (AU)

• FAPESP's process: 12/02896-9 - Prediction of RNA-Protein Interactions by an Ensemble Approach
• Grantee: Marcos Abraão de Souza Fonseca
• Support Opportunities: Scholarships in Brazil - Doctorate