*2.1. Identification of HSF Genes in M. sativa*

In this study, we used the *A. thaliana* HSF protein (AtHSF) as a query to search the *M. sativa* genome database for 25 *MsHSF* candidate genes. We then used the HMMER3 search to search the *M. sativa* genome database using the HSF-type DBD model to retrieve 22 putative *MsHSF* candidate genes (PF00447). Finally, we used SMART and NCBI conserved structural domains to delete duplicate genes and proteins without DBD conserved structural domains to obtain 16 MsHSF family members. Based on these genes' chromosomal positions, they were given the new names *MsHSF01*–*MsHSF16*.

The predicted physicochemical properties of the amino acid sequences indicated that the 16 *HSF* genes encode proteins containing 211 (MsHSF06) to 543 (MsHSF09) amino acids with molecular weights (MWs) ranging from 24,541.86 (*MsHSF06*) to 60,982.41 Da (MsHSF06), with an average molecular weight of 42,922.38 Da. The predicted isoelectric points (pI) ranged from 4.72 (MsHSF05) to 8.16 (MsHSF06), with a mean value of 5.88. Instability index calculations predicted that 14 (87.5%) of the HSF proteins were unstable in vitro, and only MsHSF08 and MsHSF12 predicted proteins with an instability index of less than 40 were classified as stable proteins. In addition, the aliphatic amino acid index (A.I.) ranged from 62.79 (MsHSF08) to 83.61 (MsHSF12), indicating that their thermal stability was less different. The overall mean of the hydrophilic (GRAVY) scores of all HSF proteins was negative, indicating that they are all hydrophilic proteins. Finally, subcellular localization predictions showed that all MsHSF proteins were located in the nucleus (Table 1).


**Table 1.** Prediction of physicochemical properties of MsHSF.
