We are uploading the phenotypic (plant height, dry biomass yield) and genotypic data files used in the publication: Novel Bayesian networks for genomic prediction of dynamic traits in biomass sorghum (dos Santos et al. 2019). Briefly, we genotyped the sorghum diversity panel using the genotyping-by-sequencing (GBS) procedure based on the PstI-HF/HinP1I and PstI-HF/BfaI restriction enzymes. The sequence reads were generated (100 bp length) on a HiSeq 4000 sequencer. Sequence reads were aligned to the Sorghum bicolor genome v3.1 (www.phytozome.jgi.doe.gov). The TASSEL 3 GBS pipeline (www.bytebucket.org/tasseladmin/tassel-5-source) was then used to call variants. The genotypic data file contain the (100,435 loci) variants already processed.