Genomic resources and draft assemblies of the human and porcine varieties of scabies mites, Sarcoptes scabiei var. hominis and var. suis

Background The scabies mite, Sarcoptes scabiei, is a parasitic arachnid and cause of the infectious skin disease scabies in humans and mange in other animal species. Scabies infections are a major health problem, particularly in remote Indigenous communities in Australia, where secondary group A streptococcal and Staphylococcus aureus infections of scabies sores are thought to drive the high rate of rheumatic heart disease and chronic kidney disease. Results We sequenced the genome of two samples of Sarcoptes scabiei var. hominis obtained from unrelated patients with crusted scabies located in different parts of northern Australia using the Illumina HiSeq. We also sequenced samples of Sarcoptes scabiei var. suis from a pig model. Because of the small size of the scabies mite, these data are derived from pools of thousands of mites and are metagenomic, including host and microbiome DNA. We performed cleaning and de novo assembly and present Sarcoptes scabiei var. hominis and var. suis draft reference genomes. We have constructed a preliminary annotation of this reference comprising 13,226 putative coding sequences based on sequence similarity to known proteins. Conclusions We have developed extensive genomic resources for the scabies mite, including reference genomes and a preliminary annotation. Electronic supplementary material The online version of this article (doi:10.1186/s13742-016-0129-2) contains supplementary material, which is available to authorized users.

The scabies mite, Sarcoptes scabiei, is an ectoparasitic acari, which causes rashes and extreme itching -known as scabies in humans. Different varieties of the scabies mite also cause mange in other species of mammals including domestic animals, livestock and wildlife. Scabies is known to cause significant morbidity in some populations, in particular Indigenous communities in Australia. We present extensive genomic sequencing data from human (Sarcoptes scabiei var. hominis) and pig (Sarcoptes scabiei var. suis) varieties of scabies mites, including Illumina whole genome sequencing data from two independent samples of adult scabies mites collected at different times from human patients from different regions of northern Australia, and from four samples of scabies mites from a pig model collected at different times and washed using different protocols to reduce bacterial contamination from host skin and mite gut. We created draft genome assemblies for var. hominis and var. suis from these resources.

Samples and sequencing
Scabies mites (var. hominis) were individually picked from skin scrapings collected 14 months apart from two unrelated patients from two different regions of northern Australia with severe crusted scabies (Patients A and B).
Over 1000 mites were collected in each sample. Two pig mange mite (var. suis) samples were collected from an inbred population of mites from a pig model [1]. The first sample consisted of >1000 mites from adult, nymph, larva and egg life stages (Pig Unwashed). The second sample, also containing all life stages, was split into three subsamples that were washed -to reduce the amount of bacteria present on the surface of the mites owing to the wound micro-environment -using three different protocols (Pig Washed 1, 2 and 3): (i) 15 min wash at room temperature in 4 % paraformaldehyde in water [2]; (ii) 1 h incubation at 37°C in 150 mM NaCl, 10 mM EDTA pH8.0, 0.6 % SDS and 0.125 μg/μl lysozyme [3]; (iii) 1 h incubation at 37°C in 1 % bleach (sodium hypochlorite) in water. In all protocols, mites were subsequently rinsed twice in water. Between wash steps, mites were centrifuged at 10,000 rpm for 2 min.
Whole mites were crushed and DNA was extracted from each sample using a QIAGEN Blood and Cell Culture DNA Kit and a modified procedure adapted from the manufacturer's protocol. Washed mites were submerged in 1 ml of ice-cold lysis buffer (20 mM EDTA, 100 mM NaCl, 1 % TritonX-100, 500 mM guanidine-HCl, 10 mM Tris pH7.9) and homogenized with stainless steel beads of 2.8 mm diameter at 6800 rpm, three cycles, 30 s per cycle, and 30 s between cycles. The suspension of lysed mites was supplemented with DNase-free RNase A to 0.2 mg/ml and with proteinase K to 0.8 mg/ml and incubated at 50°C for 1.5 h. After centrifugation at 4000 × g for 10 min to pellet insoluble debris, the genomic DNA was isolated on the QIAGEN genomic tip as per the manufacturer's protocol. Six DNA libraries were constructed and 100-base pair (bp) paired-end reads were generated using an Illumina HiSeq 2500 (see Table 1 for details).
Preliminary de novo assemblies of the adapter-and quality-trimmed reads of the Patient A, Patient B and Pig Unwashed samples were performed by using Velvet (v1.2.08) [6]. were used. The best assemblies (assessed using the scaffold N50) were obtained with a k-mer of k = 77 (Patient A, N50 = 27.4 kb), k = 63 (Patient B, N50 = 36.0 kb) and k = 81 (Pig Unwashed, N50 = 7.5 kb) (see Additional file 1 for details). Platanus (version 1.2.1) [7] was also used to perform a preliminary assembly of all six libraries, producing assemblies with better scaffold N50 values (GigaScience repository [8] for var. suis).
Since the scabies mite is a tiny, obligate parasite, it is difficult to avoid contamination from the host and from host skin and mite gut microbiomes. In addition, it was necessary to sequence thousands of intact mites, which incorporated the mite gut. Reads from the host genome were removed in silico from each sample using Bowtie 2 (version 2.2.5) [9]. Human hg19 and pig susScr3 reference genomes from the University of California, Santa Cruz, were used to build Bowtie 2 reference indices for alignment. For each sample, adapter-and quality-trimmed reads were aligned to the host reference genome using Bowtie 2 (using mode '-end-to-end' and parameter '-very-fast'). The proportion of reads aligning to host reference genomes varied from 11 to 56 % ( Table 2). Non-host reads were extracted from the alignment SAM files using the SAMtools [10] 'view' command with flag '-f 12' (read unmapped, mate unmapped).
Each host-filtered library was then assembled using Platanus (version 1.2.1, default settings), because this method performed better in the preliminary assembly of unfiltered reads. This produced assemblies with scaffold N50s ranging from 6 kb (Pig Unwashed) to 46 kb (Patient B) and major N50s up to 62 kb (see Table 2 for details). A pooled assembly of the three host-filtered washed pig samples (Pig Washed 1, 2 and 3) was also performed, producing an N50 of 4.8 kb.
The Platanus assemblies of Patient B and Pig Washed 3 had the largest major N50s (62.4 kb and 40.8 kb respectively) and were selected as the var. hominis and var. suis draft reference genomes (Table 3).
These two draft assemblies were then filtered for bacterial scaffolds by aligning scaffolds to the National Center for Biotechnology Information (NCBI) Microbial RefSeq database v72 [11] using BLASTN (version 2.2.30+; Evalue cutoff 10 −20 ; max_target_seqs = 1) [12]. The best hits in which >80 % of the scaffold length aligned to bacterial sequences were filtered out, removing 19 scaffolds from Patient B and one scaffold from Pig Washed 3. A similar search on the assemblies prior to filtering small contigs showed that most of the bacterial contigs in the assemblies were shorter than 500 bp.
To estimate the proportion of bacterial DNA contaminating the samples, microbial classification was performed on unfiltered reads from each sample using Kraken [13].  (Table 3). Protocols presented here are also available in protocols.io [14].

Estimation of genome completeness
To estimate the completeness of the assemblies, the Core Eukaryotic Genes Mapping Approach (CEGMA) [15] and Benchmarking Universal Single-Copy Orthologs (BUSCO) [16] strategies were applied to the var. hominis and suis draft genome assemblies. CEGMA (v2.5) was run with default settings on both assemblies to estimate genome completeness based on 248 ultra-conserved core eukaryotic genes (CEGs) found in nearly all eukaryotes. For both assemblies, CEGMA estimated 98.79 % completeness based on complete matches and 99.19 % completeness based on partial matches. BUSCO (v1.1b) was run in default settings using single-copy ortholog gene set databases for eukaryote taxonomic group. Seventy-five percent (75 %) of genes from the gene set of eukaryotes were predicted in both the draft genomes (66 % complete and 8.8 % fragmented genes in var. hominis and 67 % complete and 7.9 % fragmented in var. suis).

Preliminary genome annotation
A preliminary annotation of the var. hominis draft genome (Patient B) assembly was constructed by aligning UniProtKB/Swiss-Prot proteins (release 2015_07) [17] with the assembly using TBLASTN (version 2.2.30+; Evalue cutoff 10 −6 ) [12]. Multiple annotations intersecting scaffold positions on the same strand were merged into a single annotation using the BEDTools (v2.25.0) [18] 'merge' sub-command in strand-specific mode. After the merging step, a total of 13,226 gene features were annotated.

Comparison with other scabies genomics resources
The mitochondrial genome reference sequence for Sarcoptes scabiei var. hominis and var. suis have been published [19] and used to investigate within-patient diversity of infestations. A draft genome assembly of Sarcoptes scabiei var. canis is also available [20]. The scaffold N50 of this genome was 11.6 kb with a largest scaffold of 358.8 kb; the total assembly size was 56.2 Mb with a total of 18,600 scaffolds. In comparison, the var. hominis (Patient B) draft assembly had a scaffold N50 of 63.3 kb with a largest scaffold of 794.3 kb; the total assembly size was 53.6 Mb with a total of 3138 scaffolds. The annotation of the var. canis genome consisted of 10,644 predicted protein-coding genes, and the preliminary annotation of the var. hominis