Diverge and Conquer: Phylogenomics of southern Wallacean forest skinks (Genus:Sphenomorphus) and their colonization of the Lesser Sunda Archipelago

Abstract

Transcriptome Sequencing: RNA was extracted from three Sphenomorphus samples including S. striolatus (Flores, MVZ293137) and S. melanopogon from Sumbawa (MVZ293067) and Lembata (MVZ292954) using the RNEasy Protect Mini Kit (Qiagen) and provided protocol.RNA extractions were evaluated using a BioAnalyzer 2100 RNA Pico chip (Agilent). Sequencing libraries were prepared with half reactions of the TruSeq RNA Library Preparation Kit V2 (Illumina), starting with Poly-A selection for samples with high RIN scores (> 8.0). Ribo-Zero Magnetic Gold (Epicentre) was used for ribosomal RNA removal for a any samples with low RIN scores (< 8.0). After pooling the libraries they were sequenced on an Illumina HiSeq2500 with 100 bp paired-end reads. Transcriptomic sequence data were cleaned following Singhal (2013). The cleaned data were assembled using TRINITY (Grabherr et al., 2011) and annotated using the Anolis carolinensis genome (Ensembl) as a reference using reciprocal BLASTX (Altschul et al., 1997) and EXONERATE (Slater, Birney, 2005). The three annotated transcripts were compared to search for orthologs via BLAST (Altschul et al., 1990). Mitochondrial sequences were then removed and only transcripts with a GC content between 40%-70% were kept to optimize capture efficiency (Bi et al., 2012). Bioinformatics pipelines for transcriptome data processing and annotation are available at https://github.com/CGRL-QB3-UCBerkeley/DenovoTranscriptome.Marker Development: Annotated and filtered contigs from the three transcripts were aligned to identify shared exons. Markers under 300 bp were removed and those greater than 1,000 bp were cut to a maximum length of 1,000 bp. The three sets of transcripts were screened using the REPEATMASKER Web Server (Smit et al., 2015), which masks repetitive elements and low complexity regions. If any of the three transcripts for a gene contained masked sites, that gene was removed from the final marker set. The resulting 5,103 markers were compared to determine the variability of each marker in terms of the number of polymorphic sites. Invariable loci and loci with only a single variable site were discarded, as well as the top 5% of the most variable loci, leaving 3,619 candidate loci. A total of 1,199 loci were randomly chosen from the candidate loci to be integrated into an in-solution probe design for a MYBaits (MYcroarray-now Arbor Biosciences) target enrichment kit. The target size of the combined loci was approximately 1,090,000 bp. The two S. melanopogon transcripts (MVZ293067 and MVZ292954), which were ~1.6% divergent, were used as references for probe design with 3X tiling of 120bp probes across each reference. Pipelines for marker development can be found at https://github.com/CGRL-QB3-UCBerkeley/MarkerDevelopmentPylogenomics.

The effect of missing data on evolutionary analysis of sequence capture bycatch, with application to an agricultural pest

“…Developments in bioinformatic software have further enabled utilization of bycatch data, for example to detect copy number variation (Kuilman et al 2015 ; Laver et al 2022 ) from unmapped DNA and RNA reads (Zhang et al 2016 ; Gasc et al 2016 ; Laine et al 2019 )—including from public data (Vieira and Prosdocimi 2019 ). Collectively, this work demonstrates the value (and quality; Guo et al 2012 ) of sequence data derived from outside targeted regions, and its use for examining a variety of evolutionary questions is growing (e.g., Derkarabetian et al 2019 ; Ballesteros et al 2020 ; Reilly et al 2022 ; Zozaya et al 2022 ). However, while the effects of missing data in studies employing phylogenetic inference have been examined (both generally, and in the context of sequence capture; see Tilston Smith et al 2020 , and references therein), its effects on population genetic and phylodynamic analyses—particularly when the data is bycatch and therefore more likely to be patchy in nature—have received less focus.…”

“…For example, in previous human research, high-quality SNPs from outside target regions bolstered tested datasets by up to 461% (Guo et al 2012 ). Indeed, this is a growing field (e.g., Derkarabetian et al 2019 ; Ballesteros et al 2020 ; Granados Mendoza et al 2020 ; Sanderson et al 2020 ; Costa et al 2021 ; Reilly et al 2022 ; Zozaya et al 2022 ), and we recommend that more researchers consider the extraction and analysis of bycatch data (as well as other off-target genomic resources, such as unmapped RNA reads in transcriptomic studies), in their informatics pipelines. Although some of these data will undoubtedly represent contamination and/or poor quality sequences, what remains may provide the raw material for new avenues of active research (Samuels et al 2013 ; Griffin et al 2014 ; Seaby et al 2016 ).…”

“…Targeted capture, in which selected regions of the genome are sequenced following enrichment from a whole genomic DNA extract, produces sequence data that can be used to address a range of fundamental and applied biological questions (Jones and Good 2016 ), including medical (e.g., detecting disease variants; Coutelier et al 2018 ; Nagy-Szakal et al 2021 ) and eco-evolutionary (Jones and Good 2016 ). In the latter case, the resulting large multi-locus datasets are often used for phylogenomic experiments (Andermann et al 2020 ; Ballesteros et al 2020 ; Reilly et al 2022 ; Zozaya et al 2022 ), while the enrichment step makes targeted capture suitable for working with historical and ancient DNA specimens—where the available DNA is present in small amounts and in a highly degraded state (Bi et al 2013 ; Derkarabetian et al 2019 ; Roycroft et al 2022 ).…”

Integrative methods reveal multiple drivers of diversification in rice paddy snakes

“…Thus, an over-water dispersal might be the most likely dispersal scenario into Sulawesi. Many species, such as terrestrial Sphenomorphus skinks 104 , Draco lizards 105 , Cyrtodactylus geckos 106 , and frogs in the genus Limonectes 96 have been suggested to have dispersed between islands on the Lesser Sunda Arc (Lombok, Sumbawa, Komodo, Flores, and Lembata). However, although our phased concatenated tree supports a Southwest colonization of Sulawesi (likely from Selayar Island, for which samples here cluster with those of the Southwest Peninsula), a southern dispersal from the Lesser Sunda Islands is considered improbable pending the discovery of Hypsiscopus from these islands.…”

Systematics and biogeography of Lesser Sunda paddy frogs (Dicroglossidae: Fejervarya)

“…Recent genetic studies revealed that a number of wide-ranging reptile and amphibian species from the region documented solely based on morphology (e.g. Kaiser et al 2011;O'Shea et al 2012) actually comprise multiple species with localised distributions (Reilly et al 2019a(Reilly et al , 2022a(Reilly et al , 2022b(Reilly et al , 2022c(Reilly et al , 2023, suggesting that true herpetofaunal diversity of the region is underestimated.…”