A single wild male brown thornbill (B974_KIBT) from Tasmania was captured using a mist net and euthanised for genome and transcriptome generation under Australian National University Animal Experimental Ethics Committee program of wildlife authorisation approval number #A2021/33 (approval date (13/07/2021) and Tasmanian Scientific Permit #TFA23010. Every effort was made to reduce suffering of animals, including (i) collecting the minimum number of animals required for the study (one); (ii) pre-arranging animal euthanasia with a qualified veterinarian; (iii) collection of the animal from as close to the location of the veterinarian as practically possible to minimise transportation time; and (iv) transportation in a soft, dark material ‘bird bag’ to minimise stress during transportation. Tissue samples were dissected and flash frozen at -80°C or preserved in RNA later before being frozen at -80°C. High molecular weight (HMW) DNA was then extracted from kidney tissue using the Nanobind Tissue Big DNA Kit v1.0 (Circulomics). A Qubit fluorometer was used to assess the concentration of DNA with the Qubit dsDNA BR assay kit (Thermo Fisher Scientific). Total RNA was extracted from liver, brain and gonads using the RNeasy Tissue Kit (Qiagen) with RNAse-free DNAse I set (Qiagen). RNA quality was determined using the NanoDrop (Thermo Fisher Scientific) and RNA integrity (RIN) score determined using the Bioanalyzer RNA nano 6000 kit (Agilent 2100).

HMW DNA was sent for Pacific Biosciences High Fidelity (PacBio HiFi) library preparation with the SMRTbell Express Template Prep Kit 2.0 (Pacific Biosciences) and sequencing on one single molecule real-time (SMRT) cell of the PacBio Revio machine at the Australian Genome Research Facility (St Lucia, Australia). Total RNA from the liver, brain and gonads was sequenced as 150 bp paired-end (PE) reads using an Illumina Novaseq X with Illumina Stranded mRNA library preparation at the Ramaciotti Centre for Genomics (University of New South Wales, Kensington, Australia).

The genome assembly was conducted on Galaxy Australia ( The Galaxy Community, 2022) public server usegalaxy.org.au ( Afgan et al., 2016) running the Genome assembly with ‘hifiasm’ (RRID:SCR_021069) on Galaxy Australia workflow v2.1 ( Price & Farquharson, 2022). Briefly, Picard (http://broad institute.github.io/picard) (Galaxy version 2.18.2.2; RRID:SCR_006525) SamToFastq, samtools ( Danecek et al., 2021; Li et al., 2009) (Galaxy version 2.0.3; RRID:SCR_002105) flagstat and fastQC ( https://www.bioinformatics.babraham.ac.uk/projects/fastqc) (Galaxy version 0.72; RRID:SCR_014583) was used to convert BAM files to FASTQ and quality check the reads for input to Hifiasm ( Cheng, Concepcion, Feng, Zhang, & Li, 2021; Cheng et al., 2022). Hifiasm (Galaxy version 2.1) was run on Galaxy Australia to assembly the genome. Basic genome assembly statistics were calculated using the stats.sh script in BBMap v37.98 ( sourceforge.net/projects/bbmap/) (RRID:SCR_016965). Genome completeness was determined using Benchmarking Universal Single-Copy Orthologues (BUSCO; RRID:SCR_015008) v5.4.6 ( Simao, Waterhouse, Ioannidis, Kriventseva, & Zdobnov, 2015) with both the vertebrata_odb10 (n = 3354) and aves_odb10 (n= 8338) lineages on Galaxy Australia. Genome completeness and base accuracy was also determined Merqury v1.3 (RRID:SCR_022964) ( Rhie, Walenz, Koren, & Phillippy, 2020), implemented in the Genome assessment post assembly workflow on Galaxy Australia ( Price, 2023). Repetitive elements of the genome were identified, classified and masked on a Pawsey Supercomputing Centre Nimbus cloud machine (256GB RAM, 64 vCPU, 3 TB storage) by building a database using RepeatModeler v2.0.1 (RRID:SCR_015027) ( Flynn et al., 2020); repeats were then masked using RepeatMasker v4.0.9 (RRID:SCR_012954) ( Smit, Hubley, & Green, 2013-2015) with the -nolow parameter to avoid masking low complexity repeats.

The contig representing the mitochondrial genome was identified from the reference genome assembly using MitoHiFi v2 ( Allio et al., 2020; Uliano-Silva et al., 2023) and visualised using Proksee ( Grant et al., 2023). MitoHiFi identified the yellow thornbill ( Acanthiza nana) as the most taxonomically closely related publicly available mitochondrial genome (KY994614.1), used to search for the brown thornbill mitochondrial genome.

Transcriptome assembly was performed on the University of Sydney’s High Performance Computer, Artemis. Raw transcriptome reads were quality assessed pre- and post-trimming with FastQC v0.11.8 (RRID:SCR_014583). Trimmomatic v0.39 (RRID:SCR_011848) ( Bolger, Lohse, & Usadel, 2014) with the parameters SLIDINGWINDOW:4:5, LEADING:5, TRAILING:5 and MINLEN:25 and ILLUMINACLIP:2:30:10 with the TruSeq3-PE adapters was used to quality trim reads. The repeat masked genome was indexed and trimmed reads aligned using the -dta parameter with hisat2 v2.1.0 (RRID:SCR_015530) ( Kim, Paggi, Park, Bennett, & Salzberg, 2019). Resulting sam files with converted to bam format and sorted using samtools v1.9 ( Danecek et al., 2021; Li et al., 2009). Stringtie v2.1.6 (RRID:SCR_016323) ( Pertea et al., 2015) was used to generate a GTF for each transcriptome. Stringtie v2.1.6 with the -merge parameter merged transcripts into a global transcriptome retaining only transcripts with a fragments per kilobase of exon per million mapped fragments (FPKM) > 0.1 and length > 30. CPC2 v2019-11-19 ( Kang et al., 2017) was used to predict coding potential and only transcripts predicted to be coding were retained. TransDecoder v2.0.1 ( https://github.com/TransDecoder/TransDecoder) (RRID:SCR_017647) was used to predict open reading frames in the global transcriptome with a minimum transcript length of 20. Transcriptome completeness was assessed using BUSCO v5.4.6 ( Simao et al., 2015) with the vertebrata_odb10 (n= 3354) and aves_odb10 (n = 8338) lineages on Galaxy Australia.

Genome annotation was performed using FgenesH++ v7.2.2 (Softberry; RRID:SCR_018928 ( Solovyev, Kosarev, Seledsov, & Vorobyev, 2006)) using the longest open reading frame as predicted from the global transcriptome, non-mammalian settings and optimised parameters supplied with the American crow ( Corvus brachyrhynchos) gene finding matrix. BUSCO v5.4.6 ( Simao et al., 2015) in protein mode was run on Galaxy Australia to assess the completeness of the annotation with the vertebrata_odb10 (n = 3354) and aves_odb10 (n = 8338) lineages. The ‘genestats’ script ( https://github.com/darencard/GenomeAnnotation) was used to obtain the average number of exons and introns and the average exon and intron length.

The hifiasm assembly of the brown thornbill from PacBio HiFi data resulted in a genome 1.25Gb in size consisting of 1,000 contigs and sequenced to a depth of 43x. The longest contig in the assembly is 97.7 Mb and the assembly has an N50 of 20.1 Mb and L50 of 17 ( Table 1). The genome is also highly complete with 96.9% complete Aves BUSCOs present in the assembly ( Table 1). Merqury analysis also indicated a high-quality genome with QV > 59 and 87.1% complete k-mers. The mitochondrial genome is 16,862 bp and contains 37 genes including 22 tRNAs and 13 genes and 2 rRNAs ( Figure 1). Repeat masking identified 19.06% of the genome as repeats ( Table 2), which is in a similar range to other bird species ( Zhang et al., 2014).

All individual tissues had alignments rates greater than 85% against the repeat masked reference genome (liver: 93.52%, brain: 91.75% and gonads: 89.24%). A total of 45,082 transcripts were predicted to have coding potential and 12,549 longest open reading frame transcripts were used as input for genome annotation with FgenesH. A total of 29,706 genes were predicted by the FgenesH annotation software, with the annotation containing 73.9% complete aves_obd10 BUSCOs ( Table 3). There were an average number of 7.42 exons and 6.42 introns per gene ( Table 3).