A comprehensive annotation and differential expression analysis of short and long non-coding RNAs in 16 bat genomes

Nelly Mostajo Berrospi; Marie Lataretu; Sebastian Krautwurst; Florian Mock; Daniel Desirò; Kevin Lamkiewicz; Maximilian Collatz; Andreas Schoen; Friedemann Weber; Manja Marz; Martin Hölzer

Only uniquely mapped reads were counted and used for the differential gene expression analyses with DESeq2 43 (v1.16.1). Annotated rRNA genes were removed prior DESeq2 and Table 1 : We have annotated ncRNAs within 16 bat genomes of different assembly quality. We introduced threeletter abbreviations for each bat species used throughout the manuscript and in supplemental files and annotations. Genome sizes were estimated (Est.) by using C-values (DNA content per pg) from the animal genome size database (http://genomesize.com) and by applying the following formula: Genome size = (0.978 · 10 9 ) · C. If multiple entries for one species were available, an average over all C-values was calculated and used to estimate the genome size. If one species could not be found, an average C-value for the corresponding genus was used. Tab. S1 provides additional assembly statistics calculated by QUAST (v5.0.2) 24 . NCBI acc. -GenBank assembly accession without the prefix 'GCA_'. Table 2 : Six RNA-Seq data sets comprising alltogether 98 samples derived from four different bat species were used to evaluate our novel ncRNA annotations. All samples were quality trimmed and individually mapped to all 16 bat assemblies using HISAT2 34 and transcript abundances were subsequently calculated from all 1,568 mappings by featureCounts 35 . We labeld each RNA-Seq data set based on the first authors last name and the year of data set publication. Raw read data of the enriched sequencing of small RNAs (especially miRNAs) of a M. daubentonii cell line (accompanying GSE121301 36 ) have been uploaded in the course of this publication under GEO accession GSE132336 (Weber-2019). polyA+ -library preparation with mRNA selection; rRNA--library preparation with rRNA depletion and size selection (>200 nt); sRNA -library preparation with size selection (<200 nt); se/pe -single-/paired-end sequencing; ss/not-ssstrand-specific/unstranded sequencing; ss s /ss a -strand-specific in sense orientation/in anti-sense orientation. The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/738526 doi: bioRxiv preprint TPM (transcripts per million) analysis. All raw read counts from samples of one data set were combined and normalized together using the built-in functionality of DESeq2, followed by pairwise comparisons to detect significant (adjusted p-value < 0.05; absolute log 2 fold change > 2) differential expressed ncRNAs.