Integrating temporal morphophysiological and genomic markers for precise classification of flowering time in cannabis.

Indigenous Cannabis sativa populations exhibit remarkable diversity in their flowering-time responses to photoperiod cues, reflecting adaptation to varied environments. Understanding the genetic and physiological mechanisms underlying this variation is essential for optimizing cultivation and targeted breeding programs. This study characterizes flowering time diversity among 25 Iranian cannabis landrace populations using longitudinal morphophysiological and genomic approaches. Weekly data on six morphophysiological traits-stem diameter, height, growth rate, node number, internode length, and SPAD chlorophyll index-were collected over 13 weeks for female and 11 weeks for male plants, across 145 accessions genotyped by high-density genotyping-by-sequencing, yielding 233,624 high-quality SNPs. Integrated machine learning analysis of 234,002 features-encompassing SNPs, morphophysiological traits, and environmental variables-identified 53 discriminative features (22 genomic variants and 31 morphophysiological traits) that effectively classify auto, early, and late flowering accessions. Key genomic loci included AutoFlower3 (CsFTL3) on chromosome 08 and CircadianFloweringLocus1 (CsCFL1) on chromosome 09. Auto-flowering accessions showed distinct genetic and phenotypic separation consistent with photoperiod insensitivity, whereas early- and late-flowering groups exhibited overlapping distributions indicative of photoperiod-sensitive responses. These findings advance understanding of the biology of cannabis photoperiodic flowering and provide validated genomic markers for marker-assisted selection and breeding programs targeting specific photoperiod response types.