Abstract
<title>Abstract</title> <p> Background Gamma-linolenic acid (GLA) is an important omega-6 polyunsaturated fatty acid with considerable nutritional and biotechnological value; however, the transcriptional regulatory mechanisms underlying its biosynthesis in plants remain poorly understood. This study aimed to identify candidate transcription factors (TFs) and regulatory networks associated with GLA biosynthesis in the model moss <italic>Physcomitrium patens</italic> through an integrative transcriptome-based framework. Results Differential expression analysis, gene co-expression analysis, promoter motif analysis, Random Forest-based machine learning, and GENIE3 network inference were integrated to prioritize candidate regulators of three key GLA biosynthetic genes, <italic>D6D</italic> (delta-six-desaturase; <italic>Pp3c5_9590</italic> ), <italic>D12D</italic> (delta-twelve-desaturase; <italic>Pp3c1_27900</italic> ), and <italic>D6E</italic> (delta-six-elongase; <italic>Pp3c17_7080</italic> ). Four TFs representing the ERF (Pp3c6_14880), MYB-related (Pp3c12_17450), G2-like (Pp3c11_21140), and bHLH (Pp3c10_16180) families were selected for experimental validation by RT-qPCR following cold (4 and 6 h), dark, and dehydration treatments. Expression profiling largely supported the transcriptome-based predictions. The bHLH TF showed expression patterns similar to <italic>D12D</italic> , whereas MYB-related TFs exhibited expression trends consistent with D6E during dehydration. G2-like partially mirrored <italic>D6D</italic> expression, while ERF displayed expression profiles closely resembling <italic>D6D</italic> under cold stress and <italic>D12D</italic> across multiple stress conditions, supporting its proposed role as a stress-responsive hub regulator. Promoter analysis further identified <italic>cis</italic> -regulatory elements corresponding to the prioritized TF families within the promoters of their predicted target genes. Conclusions This study presents an integrative framework combining transcriptomics, machine learning, regulatory network inference, promoter analysis, and experimental validation to identify candidate regulators of GLA biosynthesis in <italic>P. patens</italic> . The results support a hub-and-module model of transcriptional regulation and identify ERF, G2-like, bHLH, and MYB-related as promising candidate regulators for future functional validation and metabolic engineering aimed at improving GLA production. </p>