Abstract

Transfer ribonucleic acids (tRNAs) and aminoacyl-tRNA synthetases (AARSs) have long been known for their indispensable roles in the translational process to synthesize proteins in living cells. Therefore, detecting and analyzing the comprehensive sets of tRNA and AARS genes would enhance understanding of the translation system and the results could potentially be used to optimize the translation efficiency for the protein production in the organisms of interest. This research aims to detect the complete sets of cytoplasmic tRNA genes and AARS genes from the nuclear genome of Myceliophthora thermophila, a filamentous fungus used in the enzyme production industry and the first thermophilic eukaryote with a finished genome sequence. In this study, 194 cytoplasmic tRNA genes and 35 aminoacyl-tRNA synthetase genes in M. thermophila were determined by comparing the gene models predicted from the genome with the sequenced RNAs and the experimentally characterized tRNA and AARS genes. The experimentally verified tRNA genes in M. thermophila can encode all of the 20 universal amino acids. The 35 AARS genes code for cytoplasmic and mitochondrial AARS enzymes of all of the 20 amino acid specificities except for the mitochondrial glutaminyl-tRNA synthetase. Four commonly used tools – tRNAscan-SE, SPLITSX, ARAGORN and tRNAfinder – were deployed for the tRNA gene prediction, and we showed that tRNAscan-SE and SPLITSX give more accurate results than ARAGORN and tRNAfinder. Analysis of the tRNA genes showed that there are significant correlations between tRNA gene number in the genome and tRNA abundance in the cell, as well as between tRNA gene number and codon usage bias of protein-encoding genes in M. thermophila. Based on the complete tRNA gene set and the codon frequencies in the protein-encoding gene set, a set of preferred codons in M. thermophila was determined. The manually curated tRNA and AARS genes of M. thermophila, as well as the experimentally characterized ones collected from tRNA and protein databases along with the supporting literature provided in this study, can be used as reliable datasets for tRNA and AARS gene annotation processes.