The genome sequence of the malaria mosquito, Anopheles funestus, Giles, 1900

We present a genome assembly from an individual female Anopheles funestus (the malaria mosquito; Arthropoda; Insecta; Diptera; Culicidae). The genome sequence is 251 megabases in span. The majority of the assembly is scaffolded into three chromosomal pseudomolecules with the X sex chromosome assembled. The complete mitochondrial genome was also assembled and is 15.4 kilobases in length.


Background
The mosquito Anopheles funestus is one of the major malaria vectors in Sub-Saharan Africa 1 . Although it can have a sparse and patchy distribution, this mosquito species is present nearly everywhere across the continent from the savannahs of West-Africa, the rainforest of Central Africa, through the dry valleys of East Africa until the Red Island of Madagascar 2 . Anopheles funestus breeds in natural and artificial, permanent or semi-permanent water bodies such swamps or rice fields. It is a member of a species group containing at least thirteen species, among which it is the most medically important species 3 . Its prominent role in the transmission of the malaria parasites is due to its close relation to humans, which provide shelters, breeding sites, and blood meals 1 . Although this association makes it highly susceptible to vector campaigns such as indoor residual spraying (IRS) and insecticide treated nets (ITNs), this mosquito species has become resistant to multiple insecticides in many parts of Africa 4 . Therefore, any program aiming at eradicating malaria cannot ignore this species.
At the genetic level, Anopheles funestus has been historically neglected in comparison to the members of the Anopheles gambiae complex. Multiple studies using genetic markers, such as microsatellites, chromosomal inversions, or DNA sequences have revealed the extraordinary genetic and inversion polymorphism of this species. This genetic richness is likely to underlie its ecological plasticity 5 , its ability to overcome insecticide pressures 6 , and incipient speciation 7 . The first complete genome draft of this mosquito appeared in 2015, originating from a colony derived from wild individuals collected in Mozambique (Fumoz) 8 . Later, the quality of the reference genome for Fumoz was improved using long read sequencing from multiple individuals and Hi-C data 9 . Here, as part of the Anopheles Reference Genomes Project (PRJEB5169), we present a chromosomally complete genome sequence for Anopheles funestus, based on a single female specimen from La Lopé, Gabon.

Genome sequence report
The genome was sequenced from a single female Anopheles funestus collected from La Lopé, Gabon (-0.187, 11.611). A total of 56-fold coverage in Pacific Biosciences single-molecule long reads (N50 10.684 kb) and 68-fold coverage in 10X Genomics read clouds were generated. Primary assembly contigs were scaffolded with chromosome conformation Hi-C data from a female sibling. Manual assembly curation corrected four missing joins or misjoins, reducing the scaffold number by 0.6%.
The final assembly has a total length of 251 Mb in 330 sequence scaffolds with a scaffold N50 of 84.637 Mb (Table 1). 92.38% of the assembly sequence was assigned to three chromosomal-level scaffolds, representing two autosomes (numbered and oriented against the AfunF3 assembly

Amendments from Version 1
The changes we made to the An. Funestus genome note are as follows  Table 2). Synteny analysis against the AfunF3 assembly revealed multiple inversions and translocations ( Figure 5), correspondence of four largest inversions to known polymorphic inversions in Anopheles funestus was revealed based on population genomics 10 and cytogenetics 11 data (Table 3), smaller inversions and other variant types will require additional validation.
The assembly has a BUSCO 5.3.2 12 completeness of 97.6% using the diptera_odb10 reference set. While not fully phased, the assembly deposited is of one haplotype. Contigs corresponding to the second haplotype have also been deposited.

Sample acquisition and nucleic acid extraction
Anopheles funestus offspring were reared from a wild caught gravid female collected from La Lopé, Gabon (latitude -0.187, longitude 11.611) by Ousman Akone-ella. A single female idAnoFuneDA-416_04 was used for Pacific BioSciences and 10x genomics, its sibling female idAnoFuneDA-416_06 was used for Arima Hi-C.
For the high molecular weight (HMW) DNA extraction for Anopheles mosquitoes, one whole insect was disrupted by manual grinding with a blue plastic pestle in Qiagen MagAttract lysis buffer and then extracted using the Qiagen MagAttract HMW DNA extraction kit with two minor  modifications. These modifications include using half volumes of the kit recommendations due to small sample size (Anopheles mosquitoes typically weigh 2-3 mg) and running two elutions of 100 µl each to increase DNA yield. The quality of the DNA was evaluated using an Agilent FemtoPulse to ensure that most DNA molecules were larger than 30 kb, and preferably > 100 kb. Single mosquito extractions ranged in total estimated DNA yield from 192 ng to 800 ng, with an average yield of 500 ng. Low molecular weight DNA was removed from using an 0.8X AMpure XP purification. A small aliquot (<~5% of the total volume) of HMW DNA was set aside for 10X Linked Read sequencing and the rest of the DNA was sheared to an average fragment size of 12-20 Kb using a Diagenode Megaruptor 3 at speeds ranging from 27 to 30. Sheared DNA was purified using AMPure PB beads with a 1.8X ratio of beads to sample to remove the shorter fragments and concentrate the DNA sample. The concentration and quality of the sheared and purified DNA was assessed using a Nanodrop spectrophotometer and Qubit Fluorometer with the Qubit dsDNA High Sensitivity Assay kit. Fragment size distribution was evaluated by running the sheared and cleaned sample on the FemtoPulse system once Sheared DNA was purified using AMPure PB beads with a 1.8X ratio of beads to sample to remove the shorter fragments and concentrate the DNA sample. The concentration of the sheared and purified DNA was assessed using a Nanodrop spectrophotometer and Qubit Fluorometer and Qubit dsDNA High Sensitivity Assay kit. Fragment size distribution was evaluated by running the sample on the FemtoPulse system once more on the sheared and cleaned sample.
For Hi-C samples, a separate sibling whole insect specimen idAnoFuneDA-416_06 was used as input material for the Arima V2 Kit according to the manufacturer's instructions for animal tissue. This approach of using a sibling was taken in order to enable all material from a single specimen to contribute to the PacBio data generation given we were not always able to meet the minimum suggested guidance of starting with > 300 ng of HMW DNA from a specimen. Samples proceeded to the Illumina library prep stage even if they were suboptimal (too little tissue) going into the Arima reaction.
To assist with annotation, which will be made available through VEuPathDB Vectorbase in due course, RNA was extracted from separate whole unrelated insect specimens idAnoFuneDA-146_02, idAnoFuneDA-367_03, and idAno-FuneDA-367_04 using TRIzol, according to the manufacturer's instructions. RNA was then eluted in 50 µl RNAse-free  water and its concentration assessed using a Nanodrop spectrophotometer and Qubit Fluorometer using the Qubit RNA Broad-Range (BR) Assay kit. Analysis of the integrity of the RNA was done using Agilent RNA 6000 Pico Kit and Eukaryotic Total RNA assay. Samples were not always ideally preserved for RNA, so qualities varied but all were sequenced anyway.

Sequencing
We prepared libraries as per the PacBio procedure and checklist for SMRTbell Libraries using Express TPK 2.0 with low DNA input. Every library was barcoded to support multiplexing. Final library yields ranged from 20 ng to 100 ng, representing only about 25% of the input sheared DNA. Libraries from two specimens were typically multiplexed on a single 8M SMRT Cell. Sequencing complexes were made using Sequencing Primer v4 and DNA Polymerase v2.0. Sequencing was carried out on the Sequel II system with 24 hour run time and 2 hour pre-extension. A 10X Genomics Chromium read cloud sequencing library was also constructed according to the manufacturer's instructions (this product is no longer available). Only 0.5ng of DNA was used and only 25-50% of the gel emulsion was put forward for library prep due to the small genome size. For Hi-C data generation, following the Arima HiC 2 reaction, samples were processed through Library Preparation using a NEB Next Ultra II DNA Library Prep Kit and sequenced aiming for 100x depth. RNA libraries were created using the directional NEB Ultra II stranded kit. Sequencing was performed by the Scientific Operations core at the Wellcome Sanger Institute on Pacific Biosciences SEQUEL II (HiFi), Illumina NovaSeq 6000 (10X and Hi-C), or Illumina HiSeq 4000 (RNAseq).

Genome assembly
Assembly was carried out with Hifiasm 13 ; haplotypic duplication was identified and removed with purge_dups 14 . One round of polishing was performed by aligning 10X Genomics read data to the assembly with longranger align, calling variants with freebayes 15 . The assembly was then scaffolded with Hi-C data 16 using SALSA2 17 . The assembly was checked for contamination as described previously 18 . Manual curation was performed using gEVAL 19 , HiGlass 20 and Pretext (https://github. com/wtsi-hpag/PretextView). The mitochondrial genome was assembled using MitoHiFi 21 , which performs annotation using MitoFinder 22 . The genome was analysed and BUSCO scores generated within the BlobToolKit environment 23 . Synteny analysis was performed with syri v1.6 24 and visualised with plotsr 0.5.3 25 .  What is the bead ratio? This information is required to make the methods complete.
Datasets clearly presented in useable and accessible format: Table 1 and links provided within the paper provide complete access to the datasets and raw sequencing reads in standard formats at standard repositories.

Is the rationale for creating the dataset(s) clearly described? Yes
Are the protocols appropriate and is the work technically sound? Yes

Are sufficient details of methods and materials provided to allow replication by others? Yes
Are the datasets clearly presented in a useable and accessible format? Yes methods and materials to allow replication by others. The datasets are clearly presented in a useable and accessible format.
Minor comments: The Fumoz genome assembly was improved not just by using long read sequencing but also by Hi-C.

○
In Table 1, Assembly accession and Accession of alternate haplotype numbers are mixed up here or on the NCBI website.

Is the rationale for creating the dataset(s) clearly described? Yes
Are the protocols appropriate and is the work technically sound? Yes

Are sufficient details of methods and materials provided to allow replication by others? Yes
Are the datasets clearly presented in a useable and accessible format? Yes