Data Collection

The lists of lncRNAs and PCGs were downloaded from the GENCODE website¹. Data from the release v27 were used for human genes annotated on the genome sequence GRCh38 (gencode.v27.long_noncoding_RNAs.gtf.gz; gencode.v27.basic.annotation.gtf.gz). Data from the release M16 were used for mouse genes annotated on the genome sequence GRCm38 (gencode.vM16.long_noncoding_RNAs.gtf.gz; gen code.vM16.basic.annotation.gtf.gz). PCGs were selected from the basic annotation when both gene and transcript were indicated as “protein_coding”. The total number of genes, transcripts and exons considered in both species are reported in Supplementary Table S1.

An independent validation of the results from GENCODE was obtained by collecting human lncRNAs annotations data from 6 different databases: the FANTOM5 database (Fantom CAT genes²; FANTOM_CAT.lv3_robust.only_lncRNA.gtf) (Hon et al., 2017), the NONCODE v.5 database³ (Fang et al., 2018), the BIGTranscriptome database release 2016 lncRNA catalog⁴ (You et al., 2017), the LncBook database⁵ (Ma et al., 2019), the MiTranscriptome database⁶ (Iyer et al., 2015), and the LNCipedia database version 5.2⁷ (Volders et al., 2013). A validation of results obtained from the mouse genome was performed using lncRNAs annotations from the NONCODEv5 database⁸.

The lists of lncRNAs and PCGs of Drosophila melanogaster and Caenorhabditis elegans were downloaded from the BioMart data mining tool (Smedley et al., 2015) in the Ensembl genome database (release 91).