Banca de QUALIFICAÇÃO: RODRIGO LIMA ANDRADE

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : RODRIGO LIMA ANDRADE
DATE: 20/07/2026
TIME: 14:00
LOCAL: Google Meet
TITLE:

GTx-GUT: An Automated Pipeline for the Analysis and Clinical Interpretation of Human Gut Microbiome Data

KEY WORDS:

Gut microbiome; 16S rRNA sequencing; bioinformatics; automation; pipeline; artificial intelligence; QIIME 2; Snakemake

PAGES: 53
BIG AREA: Ciências Biológicas
AREA: Biologia Geral
SUMMARY:

The gut microbiome plays a fundamental role in several physiological processes of the host and is associated with different pathological conditions, including obesity, type 2 diabetes, inflammatory bowel diseases, and neuropsychiatric disorders. In this context, microbiome sequencing data analysis has become a relevant approach for characterizing microbial communities and investigating their impacts on human health. However, the bioinformatics processing of such data involves multiple steps, tools, and parameters, which hinders analytical standardization, compromises reproducibility, and makes result interpretation more complex. In this scenario, this work presents the development of GTX-GUT, an automated pipeline designed for gut microbiome analysis based on 16S rRNA gene sequencing data. The system integrates, within a single and reproducible workflow, quality control with BBDuk, bioinformatics processing with QIIME 2 and DADA2, taxonomic classification with GreenGenes 13.8, alpha diversity metrics calculation, statistical comparison against a reference population of 530 healthy individuals, clinical association analysis, and automated report generation in Word format. The pipeline also incorporates artificial intelligence support through GPT-4o and GPT-4o-mini models for textual synthesis and interpretation of findings. Functional validation with a ZymoBIOMICS mock community demonstrated that the preprocessing step reduces distortions in taxonomic classification. Application to a real human sample produced a complete clinical report, with a Shannon index of 3.264, an F/B ratio of 0.36, classification as Enterotype 1, and identification of bacterial associations across 11 clinical categories. GTX-GUT represents a methodological contribution to bioinformatics applied to the gut microbiome, combining computational automation, analytical standardization, and support for scientific interpretation in a reproducible workflow orchestrated by Snakemake and executed within Singularity containers.

COMMITTEE MEMBERS:
Interno - 2170415 - JORGE ESTEFANO SANTANA DE SOUZA
Presidente - 1939184 - SANDRO JOSE DE SOUZA
Interno - 3063244 - TETSU SAKAMOTO