Banca de QUALIFICAÇÃO: PEDRO VICTOR BARBOSA ARAUJO

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : PEDRO VICTOR BARBOSA ARAUJO
DATE: 19/02/2026
TIME: 14:30
LOCAL: Google Meet
TITLE:

Development of an automated computational pipeline for generating polygenic risk score reports.

 

KEY WORDS:

Polygenic risk scores; PRS; PRS report; phenotypic traits; genetic variants; precision medicine; GWAS; PLINK2; LangFlow; LaTeX

 

PAGES: 45
BIG AREA: Ciências Biológicas
AREA: Biologia Geral
SUMMARY:

Polygenic Risk Scores (PRS) have emerged as a promising approach for assessing the genetic risk of complex diseases by aggregating the effects of multiple genetic variants associated with a given phenotype. Despite their potential, the practical application of PRS still faces challenges related to methodological standardization, clinical interpretation of results, and the integration of these data into reports that are accessible to healthcare professionals. In this work, an automated computational pipeline is proposed and implemented for the generation of PRS reports, encompassing data acquisition and genomic data processing through to the structured presentation of results and the generation of personalized clinical recommendations for multiple phenotypic traits, enabling the simultaneous analysis of different conditions and complex characteristics within the same individual. The pipeline relies on publicly available data from genome-wide association studies (GWAS) and the 1000 Genomes Project as a reference population, employing PLINK2 for score calculation, result normalization, and the derivation of comparative metrics such as z-scores and percentiles. Beyond PRS calculation, the system performs phenotype categorization, automatic gener- ation of graphical visualizations, and individual risk interpretation across different risk levels. The clinical recommendation generation stage is implemented using LangFlow, which orchestrates language models to interpret each phenotype based on its description and estimated risk level. Recommendations are pre-structured and stored in JSON format, ensuring reusability, consistency, and seamless integration into the final LaTeX-generated report. The results demonstrate that the proposed pipeline enables the production of reproducible, interpretable, and integrable PRS reports, contributing to the systematic use of PRS as a decision-support tool in precision medicine while acknowledging its methodological and clinical limitations.

 

COMMITTEE MEMBERS:
Interna - 1365498 - BEATRIZ STRANSKY FERREIRA
Interno - 1267860 - GUSTAVO ANTONIO DE SOUZA
Presidente - 1939184 - SANDRO JOSE DE SOUZA