UCE phylogenomics reveals a new tribe: Gondwanocentrini

UCE data processing:Sequencing raw reads were filtered and trimmed with Illumiprocessor (Faircloth 2013; Bolger et al. 2014), available in phyluce v1.7.1 (Faircloth 2016) and assembled with SPAdes version 3.15.5 (Bankevich et al. 2012). The resulting contigs were then processed following the Phyluce v1.7.1 pipeline (Faircloth 2016). Contigs were first queried against a FASTA file of all enrichment baits, creating a relational database with the location of the UCE loci. Samples that recovered &lt;100 UCE loci were discarded from the pipeline and not used in downstream analyses. Individual loci were then extracted and placed in separate FASTA files, and each locus was aligned using MAFFT v. 7.130b (Katoh et al. 2002; Katoh and Standley 2013) and trimmed with GBLOCKS v. 0.91b (Castresana 2000; Talavera and Castresana 2007) with reduced stringency settings (0.5, 0.5, 12, and 7 for b1–b4 settings, respectively). Alignments were filtered with different settings to produce two matrices with different levels of completeness, one with loci available for at least 70% and the other for at least 50% of the terminal taxa. All the bioinformatic processing and phylogenetic analyses were conducted in the High-Performance Computing server BEAGLE (Dell Power Edge R740) owned by IBUNAM, except for the sequence assembly, which was performed in the Galaxy public server (usegalaxy.eu)28s+COI sequences:Matrix data was assembled including sequences from previous studies plus new sequences, including the new taxa, from the raw sequence reads that were originally generated for UCE data using the "match_contigs_to_barcodes" tool available in Phyluce v1.7.1 (Faircloth 2016), using the 28S and COI sequences of <i>Conobregma</i> sp. 3 as templates. Only taxa with &gt; 1000 called bases were included as some with fewer bases were recovered far from expected relatives in preliminary analyses. Alignment of COI was trivial as there were no indels. The length-variable 28S sequences were aligned according to the secondary structure model of Gillespie et al. (2005) as in other studies (Butcher et al., 2014; Quicke et al., 2016). For the 28S gene, only confidently alignable positions were included in the analyses. Data were partitioned into the three COI codon positions and pairing and non-pairing bases of the RNA gene (Quicke et al. 2020, 2021).<br>The matrices added are UCE 50 and 70%, and 28S+COI. Additional files define the partition and secondary structure used for the 28S+COI matrix.