Long read genome assembly of Automeris io (Lepidoptera: Saturniidae) an emerging model for the evolution of deimatic displays

Automeris moths are a morphologically diverse group with 145 described species that have a geographic range that spans from the New World temperate zone to the Neotropics. Many Automeris have hindwing eyespots that are thought to deter or disrupt the attack of potential predators, allowing the moth time to escape. Some species in the genus have vestigial eyespots or lack them completely, suggesting that this trait may provide a selective benefit. The Io moth (Automeris io), known for its striking eyespots, is the most widely studied species within the genus and is an emerging model system to study the evolution of deimatism, a predatory defense that combines visual stimuli and movement. Here we present a high-quality, PacBio HiFi genome assembly for Io moth to aid existing research on the molecular development of eyespots. Genomic research is needed to address questions involving antipredatory defenses and eyespot pattern development. BUSCO analysis for this genome shows a completeness of 98.4%, and N50 of 15.

Methods
PacBio consensus reads were used to estimate genome size and heterozygosity using the program K-Mer counter (KMC) v.3.2.1 (RRID: SCR_001245). A k-mer length of 23 (-m 23) was used to create a histogram of k-mer frequencies and visualized using GenomeScope 2.0 (RRID:SCR_017014). PacBio consensus reads were assembled using the de novo assembler, HiFiasm v.0.16.1 r307 (RRID:SCR_021069), resulting in a 500 Mbp primary assembly. The assembly_stat.py script was used to assess assembly contiguity. BUSCO v.5.2.0 was used to assess completeness with 5,286 putative single copy genes from the lepidoptera_odb10.2019-11-20 database (RRID:SCR_015008; ). Therefore, we attempted to further collapse allelic variation using the Purge Haplotigs pipeline was used to purge additional duplicates. (purge_haplotigs v.1.1.2 ). A coverage histogram was used to choose a minimum, median, and maximum read depth cut off value for purging. This was produced by mapping raw reads to the primary assembly using minimap v.2.21 (RRID:SCR_018550). Contigs were assigned as haplotigs if the 80% of the contig showed diploid-level coverage (-s 80) and discarded if coverage was 80% above or below the read depth cut offs (-j 80). We used blobtools v1.0 (RRID:SCR_017618) to investigate potential contamination in the assembly. RepeatMasker  was used to mask repeat regions of the assembly, creating a soft-masked genome to be used for all downstream analyses. The BRAKER2 pipeline (v2.1.5; RRID:SCR_018964) with protein sequences from the NCBI Bombyx mori Annotation Release 101 was used for structural annotation. This pipeline uses programs Bamtools (RRID:SCR_015987;, GeneMark-EP+ (RRID:SCR_011930, DIAMOND (RRID:SCR_016071), and Augustus (RRID:SCR_008417). Annotation statistics were summarized using gFACs (RRID:SCR_022017. BUSCO v.5.2.0 was used to assess completeness of this annotation. Refer to original publication for further details.