Gossypium raimondii (D5) 'D5-4' genome NSF_v1

Material and Methods

DNA was extracted from G. raimondii (accession D5-4) plants using CTAB techniques (Kidwell and Osborn 1992). DNA concentration was measured by a Qubit Fluorometer (ThermoFisher, Inc.). The sequencing library was constructed according to PacBio recommendations at the BYU DNA Sequencing Center (DNASC). Fragments >18 kb were selected for sequencing via BluePippen (Sage Science, LLC). Prior to sequencing, the size distribution of fragments in the libraries was evaluated using a Fragment Analyzer (Advanced Analytical Technologies, Inc). Eight and eleven PacBio cells were sequenced from a single library for G. raimondii, respectively, on the Pacific Biosciences Sequel system. For both genomes, the raw PacBio sequencing reads were assembled using Canu V1.6 using default parameters (Koren et al. 2017).

HiC libraries were constructed from G. raimondii leaf tissue at NorthEast Normal University, China. Sequencing was performed at Annoroad Gene Technology Co., Ltd (Beijing, China). The HiC data of G. raimondii was mapped to the previous genome sequence of G. raimondii using HiC-Pro (Servant et al. 2015), and to the newly assembled CANU contigs of G. raimondii PacBio reads by PhaseGenomics. The HiC interactions were used as evidence for contig proximity and in scaffolding contig sequences. An initial draft genome sequence of pseudochromosomes (PGA assembly) was created using a custom python script from PhaseGenomics.

DNA was also extracted from young G. raimondii leaves following the Bionano Plant protocol for high-molecular weight DNA. DNA was purified, nicked, labeled, and repaired according to Bionano standard operating procedures for the Irys platform. Two optical maps of different enzymes (BspQI and BssSI) were assembled using the IrysSolve pipeline on the BYU Fulton SuperComputing cluster. The optical maps were combined into a two-enzyme composite optical map and it was aligned to the PGA assembly using an in silico labeled reference sequence. Conflicts between the Bionano maps and the PGA assembly were manually identified in the Bionano Access software by comparing the mapped Bionano contigs to the CANU contigs along the draft genome sequence. Conflicts between datasets were resolved by repositioning and reorienting CANU contigs in PGA ordering files followed by reconstruction of the fasta sequence, provided there was supporting or no-conflict evidence from the optical map (Durand et al. 2016). Multiple iterations of mapping, conflict resolution, and draft sequence construction resulted in the final, new genome sequence of G. raimondii.

About the assembly

The G. raimondii genome was assembled from 43.7x PacBio coverage of raw sequence reads. The assembly consisted of 187 contigs with an N50 of 6.3MB (Table 1). The contigs were
scaffolded using HiC by PhaseGenomics and the pseudomolecules were manually adjusted using JuiceBox (Durand et al. 2016). The final scaffolded assembly was independently verified using a composite optical map of two different enzymes. A comparison of assembly metrics between the previous genome sequence and our new genome sequence of G. raimondii illustrates a 45x improvement in contig length and a 97x reduction in the number of gaps. The cumulative gap length of the new assembly (17.6 kb) was reduced by 647x compared to the assembled gaps of the previous genome sequence (11,391 kb). The final genome assembly size was 14.9 MB smaller than the previous assembly, representing 98% of previously assembled genome sequence in length.

Table 1. Assembly metrics of the G. turneri genome, the G. raimondii (our current assembly, D5), and the previous G. raimondii assembly (Paterson et al. 2012).

Publication

Udall, et al., 2019. De novo genome sequence assemblies of Gossypium raimondii and G. turneri.