<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<ArticleSet>
<Article>
<Journal>
<PublisherName>International Academy of Ecology and Environmental Sciences</PublisherName>
<JournalTitle>Ornamental and Medicinal Plants</JournalTitle>
<eissn>2522-3682</eissn>
<Volume>9</Volume>
<Issue>1-4</Issue>
<PubDate PubStatus="ppublish">
<Year>2026</Year>
<Month>9</Month>
<Day>1</Day>
</PubDate>
</Journal>
<ArticleTitle>Fetching GWAS (Genome-Wide Association Study) data via AI: A web 
tool to synthesize genotype, phenotype, and summary statistics</ArticleTitle>
<Pages>1-21</Pages>
<Language>EN</Language>
<AuthorList>
<Author>WenJun Zhang</Author>
<Author>Yanhong Qi</Author>
</AuthorList>
<ArticleList>
<ArticleId IdType="url">http://www.iaees.org/publications/journals/omp/articles/2026-9(1-4)/fetching-GWAS-data-via-AI.pdf</ArticleId>>
</ArticleList>
<Abstract>
A GWAS (Genome-Wide Association Study) data fetcher via AI was developed in present study. It is a web tool that generates realistic synthetic GWAS data based on user inputs via AI APIs (DeepSeek, Google Gemini, or OpenAI GPT). It outputs three data components: (1) summary statistics: an array of SNP records (CHR, SNP, BP, A1, A2, FRQ_A1, BETA, SE, P, N, INFO), (2) phenotype data: an array of individual records (FID, IID, PHE, SEX, AGE, PC1, PC2, BATCH), and (3) metadata: a descriptive string containing genotyping platform, QC protocols, genome build, and population notes. By leveraging AI to generate realistic data, users can practice analysis without accessing restricted genetic databases, test analysis pipelines safely, learn GWAS data structure and format, and develop and validate bioinformatics tools.
</Abstract>
</Article>
</ArticleSet>
