Fig. 7
From: Harnessing large language models (LLMs) for candidate gene prioritization and selection
Transcriptional profiles of the top 5 scoring candidate genes across 16 reference blood transcriptome datasets. A The box plot shows the changes in abundance of transcripts comprising the M9.2 module, expressed within a cohort as the percentage of genes showing differential expression between each patient and the respective set of control subjects for this cohort. Differential expression is based on a fixed cut-off (Log2 Fold Change > 1.5, difference > 10). These values are expressed as % module activity and can range from + 100% (all genes are upregulated) to -100% (all genes are down regulated). B Box plots show expression levels for the top 5 scoring M9.2 genes in patient cohorts relative to respective control subjects matched for demographics variables. The datasets were contributed by Altman, Rinchai et al. (GSE100150) [17]