Reptilian Transcriptomes v2.0: An Extensive Resource for Sauropsida Genomics and Transcriptomics

Abstract

Despite the availability of deep-sequencing techniques, genomic and transcriptomic data remain unevenly distributed across phylogenetic groups. For example, reptiles are poorly represented in sequence databases, hindering functional evolutionary and developmental studies in these lineages substantially more diverse than mammals. In addition, different studies use different assembly and annotation protocols, inhibiting meaningful comparisons. Here, we present the “Reptilian Transcriptomes Database 2.0,” which provides extensive annotation of transcriptomes and genomes from species covering the major reptilian lineages. To this end, we sequenced normalized complementary DNA libraries of multiple adult tissues and various embryonic stages of the leopard gecko and the corn snake and gathered published reptilian sequence data sets from representatives of the four extant orders of reptiles: Squamata (snakes and lizards), the tuatara, crocodiles, and turtles. The LANE runner 2.0 software was implemented to annotate all assemblies within a single integrated pipeline. We show that this approach increases the annotation completeness of the assembled transcriptomes/genomes. We then built large concatenated protein alignments of single-copy genes and inferred phylogenetic trees that support the positions of turtles and the tuatara as sister groups of Archosauria and Squamata, respectively. The Reptilian Transcriptomes Database 2.0 resource will be updated to include selected new data sets as they become available, thus making it a reference for differential expression studies, comparative genomics and transcriptomics, linkage mapping, molecular ecology, and phylogenomic analyses involving reptiles. The database is available at www.reptilian-transcriptomes.org and can be enquired using a wwwblast server installed at the University of Geneva.

The genome sequence of the corn snake (Pantherophis guttatus), a valuable resource for EvoDevo studies in squamates

“…We have long supported the use of this species as a model for evolutionary developmental biology in snakes (Brykczynska et al, 2013, Di-Poi et al, 2010, Milinkovitch and Tzika, 2007, Tzika et al, 2011, Tzika and Milinkovitch, 2008, Tzika et al, 2015. We are particularly interested in understanding the genetic basis of colour and colour pattern variation within and among snake species.…”

“…MAKER2 provides a set of gene predictors, performs similarity searches against cDNA/ESTs or proteins databases, and combines the outputs into a set of highquality gene predictions. Our corn snake transcriptomic data (Tzika et al, 2015) together with a protein database (including the SwissProt repository and NCBI sequences of A. carolinensis and snakes) were used in the MAKER2 pipeline to complement the gene models computed by three ab-initio/ evidence-based predictors: AUGUSTUS (Stanke and Waack, 2003), SNAP (Korf, 2004) and GeneMark-ES (Ter-Hovhannisyan et al, 2008). As a quality measurement of our annotation, we used the 'annotation edit distance' (AED, (Eilbeck et al, 2009)), which is based on the agreement at the nucleotide level (i.e., focusing on overlapping nucleotides rather than substitutions) between the final gene annotation and the aligned evidence data (cDNA/proteins) supporting that annotation.…”

“…These rather low numbers are due to the high fragmentation of the draft corn snake genome: many genes span multiple scaffolds, making their annotation impossible. Eighty percent (8,817) of the corn snake proteins with AED < 1 had a blastp hit (e-value <10 -5 ) against A. carolinensis (Ensembl v77 proteins), P. molurus or O. hannah sequences and 94% (10,237) had a megablast hit against our P. guttatus transcriptome (Tzika et al, 2015). When considering proteins with the maximum AED value of 1, only 19.2% (2,565) had a hit against A. carolinensis, P. molurus or O. hannah and 39.4% (5,263) against the P. guttatus transcriptome.…”

“…For the gene annotation, we built a protein database including the SwissProt database, as well as all A. carolinensis and snake proteins (including the ones from P. molurus and O. hannah) from NCBI. We also used a P. guttatus transcriptome assembled from Illumina and 454 cDNA libraries obtained from a mix of adult organs (testis, kidneys, brain and vomeronasal organ) and three developmental stages (Brykczynska et al, 2013, Tzika et al, 2015.…”

Convergent origination of aDrosophila-like dosage compensation mechanism in a reptile lineage

“…One-to-one orthologous sequences were obtained from the Ensembl database (version 81; www.ensembl. org) and the Reptilian Transcriptomes v1.0 and v2.0 database (Tzika et al 2011(Tzika et al , 2015. We used three phylogenetic methods to assess the origin of seven Y/X gametologs and the concatenated stratum 1 gene set (RPL6 and PPP1CC): (1) A Bayesian-based phylogenetic reconstruction using MetaPIGA (Helaers and Milinkovitch 2010) with the GTR model but without invariable sites, 110 replicates with 200 iterations each, trees sampled every five generations.…”