brunoasm / Labrundinia_phylo

Code used to process Illumina data for Labrundinia phylogeny

Anchored hybrid enrichment using PCR-based probes for chironomid phylogenies

This repository includes code used to assemble genes sequenced with anchored hybrid enrichment. For more information, check the publication:

Silva, F.L., de Medeiros, B.A.S., & Farrell, B.D. (2024). Once upon a fly: The biogeographical odyssey of Labrundinia (Chironomidae, Tanypodinae), an aquatic non-biting midge towards diversification. Molecular Phylogenetics and Evolution, 108025. https://doi.org/10.1016/j.ympev.2024.108025

Modifications to standard SECAPR assembly

This dataset presents some complications that required modifications from standard SECAPR assembly:

1 - abyss cannot assemble large contigs, likely due to high heterozygosity Solution: use redundans to refine assemblies

2 - genera sequenced have a large divergence and no close references for each one Solution: extract contigs in two rounds:

First with a single similarity threshold for all samples and a single reference Second, choose references from extract genes, blast them on NCBI to validate, and redo extraction independently for each genus

Reference sample used in second round:

	EF1A	CAD1	CAD4	AATS	PGDI	TPI
Ablabesmyia	23X1	23X1	23X1	23X1	23X1	45X1
Chironomus	29X1	31X22	31X22	31X22	31X22	31X22
Labrundinia	4X1	4X1	4X1	31X9	4X1	4X1
Polypedilum	42X6	42X6	42X6	42X6	42X6	42X6

I manually concatenated the references for each genus in a fasta file. These are stored in the folder extracted_by_genus as reference_$GENUS.fasta

3 - There are ribosomal and mitochondrial genes These were assembled with mitobim instead of abyss, and assemblies concatenated to nuclear protein coding genes for alignment

4 - coverage for some samples is too low after duplicate removal and SECAPR halts To overcome this limitation, mapping and consensus was done in 2 steps: First, mapping without removing duplicates Based on this, mannually removing sequences with low coverage from alignments These new alignments are in folder contig_alignment_filtered Then redoing mapping, this time removing duplicates

These are the genes for final alignments:

ID	NAME
0	Tanytarsus_EF1alpha
1	Tanytarsus_CAD1
2	Tanytarsus_CAD4
3	Tanytarsus_AATS
4	Tanytarsus_PGDI
5	Tanytarsus_TPI
6	18S
7	28S
8	16S
9	COI_COII