Extended Data 3: tRNAseq analysis of Blastocrithidia nonstop

Individual alignments corresponding to each identified tRNA species in Blastocrithidia nonstop.For tRNA sequencing, Illumina sequencing of cDNA of cytosolic RNA fraction corresponding to tRNAs was used. Cytosolic RNA fraction was obtained as described previously (Paris et al. 2021) and submitted to Arraystar, Inc. (Rockville, MD, USA) for tRNA sequencing analysis. Before the sequencing experiment, the integrity and quantity control of each RNA sample using agarose gel electrophoresis and the NanodropTM instrument was performed. For tRNA isolation over 2μg total RNA was resolved on urea-polyacrylamide gels and recovered within a size window of 60-100 nt for tRNA followed by coupled nucleotide demethylation and Hydro-tRNAseq method to further improve the efficiency of tRNA-seq library preparation, containing the following treatments: a) m1A and m3C demethylation of purified tRNA; b) limited alkaline hydrolysis of demethylated tRNA by carbonate buffer; c) calf intestinal phosphatase (CIP) dephosphorylation and then T4 polynucleotide kinase (T4 PNK) re-phosphorylation of the partially hydrolyzed tRNA fragments. Fragments of ∼19-35nt for partially hydrolyzed and re-phosphorylated tRNA were converted to small RNA sequencing libraries using NEBNext® Multiplex Small RNA Library Prep Set for Illumina® kit.The libraries were qualified and absolutely quantified using Agilent BioAnalyzer 2100. For tRNA sequencing on Illumina NextSeq 500 instrument, the sequencing type was 50-bp single-read.The Illumina sequencing yielded 7,912,517 reads with length of 50 nt. The reads were adapter- and quality-trimmed and mapped on the B. nonstop genome assembly to filter out low-quality and contaminating sequences. The 77 % of the trimmed reads that were successfully aligned on the assembly were used for further analysis.Firstly, vsearch v2.5.2 (Rognes et al., 2016) was used to cluster identical reads. 22,436 read clusters were obtained. To identify the origin of the sequences, clusters were aligned to B. nonstop tRNA and rRNA gene database using blastn from BLAST v.2.3.0+ (here and later in the text) (Camacho et al., 2009). The clusters together with their best BLAST hit (i.e., corresponding tRNA) were split into individual fasta files, and each dataset was aligned using MAFFT with “ --localpair --op 10 --ep 0 --maxiterate 1000” parameters. Each of the resulting alignments was manually analysed in Aliview (Larsson, 2014).