Workflow científico de anotação genômica funcional e curadoria manual de genomas

Abstract

With the advent of new generation sequencing technologies, genome sequencing was no longer a scientific barrier. Analysing the structure and giving biological meaning to genomic and proteomic sequences in silico(genome annotation) became, nevertheless, the new challenging task. Functional annotation is a crucial step in this process and aims to understand organisms? biological processes and guide new research. However, it is a complex task, as it involves usinga large set of databases, web servers, and tools to compare all the sequences of interest with other sequences available in protein domain repositories. Each of these tools has independentand format-specific output files, making it difficult to keep annotation data organized in a single-context environment and preventing researchers from working collaboratively within a project. This research aimed to develop a scientific workflow for functional annotation and manual curation of genomes by implementing a collaborative web application that supports the automatic execution of functional annotation tools and integrates results into a single-context platform. The solution also supports manual curation of structural and functional data for each gene and allows users to export annotation results to file formats required by GenBank database?s submission process. The workflow was tested and validated in the annotation process of wild-type (2HH) and mutant (S1M29) strains of Penicillium echinulatum, conducted by Computational Biology and Bioinformatics Laboratory at University of Caxias do Sul and submitted to GenBank under BioProjects PRJNA520890 and PRJNA521489, respectively. Besides, an experimental version of the tool was made freely available at https://seq2annot.org. The solution has proven effective as it abstracts the complexity of the external tools? execution and integrates annotation results in a unified collaborative environment. The core mechanism of asynchronous task execution provided the ability to horizontally scale server resources as demand grows over time. The architectural decision of splitting jobs into single independent tasks also ensured that new functional annotation tools can be easily developed and plugged into the application, making it a promising tool to be continuously improved and used in large-scale genome annotation projects in the future. [resumo fornecido pelo autor]