/genomeassembly-pacbiohifi

pacbioHifi genome assembly benchmarks

Primary LanguageHTMLMIT LicenseMIT

genomeassembly-pacbiohifi

I completed analysis and benchmarking of the pacbiohifi for the genome assembly coming from the PacBioHifi reads for the genotypes and haplotypes and their quality assessment. This is the complete genome assembly standardization coming from the PacBioHifi reads and includes all the scripts and the analysis for the reproducibility and the visualziation and the summary analysis.

Genome assembly packages used for the assembly

Genome assembly evaluation

Genome Analysis and Project Summary

image

image

Haplotype based assembly

  • ERR10930361.fastq - maternal
  • ERR10930362.fastq - paternal
  • ERR10930363.fastq -child
  • The hifiasm generates two files when you do the trio binning based on the maternal and the paternal and the filenames outputted are maternalpaternal.asm.dip.hap1.p_ctg.fa and maternalpaternal.asm.dip.hap2.p_ctg.fa and hence they are long so i put the hap1 for the first one and the hap2 for the second one.
  • The hifiasm trio binning results are present at trio binning
  • All the configuration, run time, wallclock time, nodes, cpu, memory are listed here along with the runtime for each of the analysis project parameters
haplotypes contigs Largest contig Total length GC (%) N50 N75 L50 L75
hap 1 823 37880845 534683615 35.47 15838561 9224513 12 23
hap 2 116 31333120 502852466 35.28 23599708 18852907 10 16
  • BUSCO evaluation of both the haplotypes: completeness based on BUSCO for the trio binning method ( Total BUSCO searched 425)
Species Complete BUSCO Complete and single-copy BUSCO Complete and duplicated BUSCOs (D) Fragmented BUSCOs (F) Missing BUSCOs (M) Completion score
hap1 422 415 7 1 2 99.2%
hap2 420 407 13 4 1 98.9%

Haplotype assembly without the trio method:

  • The last l3 means that for the same sample, the -l3 extensive purging was involved.
contigs ERR10930361 ERR10930362 ERR10930363 ERR10930364 ERR10930364_l3
1248 2593 2611 1640 312
  • since the name of the files after the ragtag are so long so i have named them like this for the easier read:
    • maternalpaternal.asm.dip.hap2.p_ctg.fa.ragtag.correct.scaffold: ragtag1
    • maternalpaternal.asm.dip.hap1.p_ctg.ragtag.correct.scaffold: ragtag2

Ragtag assembly stats on the tribinning asm methods.

haplotype name placed_sequences placed_bp unplaced_sequences unplaced_bp gap_bp gap_sequences
ragtag1 293 473623733 1619 61059882 27100 271
ragtag2 302 465201572 1054 37650894 28100 28

Quast results after the scaffolding with the ragtag.

  • The number of the scaffolds have increased and this might be due to the fact that the genome is polyploid so the scaffold have been placed accordingly. The assembled size is near to the grapevine genome.
haplotypes contigs Largest contig Total length GC (%) N50 N75 L50 L75
ragtag1 1626 37075411 534706754 35.47 24081286 20838642 10 16
ragtag2 1050 35080446 502871947 35.28 23258691 21132941 10 15

BUSCO report for the ragtag corrected scaffolds.

Species Complete BUSCO Complete and single-copy BUSCO Complete and duplicated BUSCOs (D) Fragmented BUSCOs (F) Missing BUSCOs (M) Completion score
ragtag1 422 415 7 1 2 99.2%
ragtag2 421 408 13 3 1 99.1%

Verkko assembly has been finished, parameters added to the parameters, including the wallclock time, resources used, time of execution, summary below: kmer used 30

  • assembly stats of the verkko assemblies using the trio binning. kmer used 30
haplotypes contigs Largest contig Total length GC (%) N50 N75 L50 L75
verkko hap 1 2259 33896910 508997582 35.56 10831024 5617093 16 31
verkko hap 2 995 26795713 477196271 35.13 13168430 7656591 13 23
  • assembly stats of the verkko assemblies using the trio binning. kmer used 50
haplotypes contigs Largest contig Total length GC (%) N50 N75 L50 L75
verkko hap 1 2511 33896910 524953580 35.53 10586138 5279585 17 34
verkko hap 2 1061 26317445 485115873 35.14 13168430 7577555 13 24
  • BUSCO results from the verkko assemblies using the trio binning. kmer used 30
assembly type complete busco complete and single copy busco complete and duplicated busco fragmented busco missing busco completion score
hap1 393 385 8 1 31 92.5%
hap2 394 384 10 3 28 92.8%
  • BUSCO results from the verkko assemblies using the trio binning. kmer used 50
  • By using the high kmer the BUSCO score improved.
assembly type complete busco complete and single copy busco complete and duplicated busco fragmented busco missing busco completion score
hap1 401 392 9 0 24 94.3%
hap2 400 389 11 3 21 94.1%

  • assembly mapping reads using the pacbiohifi mapping and the coverage for the haplotypes.

    -- bamstats coverage for hap1 and hap2

    Haplotype Total reads Mapped reads Forward strand Reverse strand
    hap1 3866068 3863435 (99.9319%) 1936181 (50.0814%) 1929887 (49.9186%)
    hap2 3656794 3653759 (99.917%) 1831768 (50.0922%) 1825026 (49.9078%)

    -- bamstats for the hap1 and hap2

    haplotype total unmapped mapped mappings ratio mappings count
    hap1 2330837 2633 2328204 1.65941 3863435
    hap2 2529319 3035 2526284 1.4463 3653759

Genome assembly stats for the hifiasm after polishing the genome with the ERR10930363.fastq reads

  • QUAST reports
haplotypes contigs Largest contig Total length GC (%) N50 N75 L50 L75
hap 1 2259 33896900 508823952 35.56 10831023 5617076 16 31
hap 2 995 26781654 477006706 35.13 13168428 7656584 13 23
  • BUSCO reports of hifiasm polished genome
assembly type complete busco complete and single copy busco complete and duplicated busco fragmented busco missing busco completion score
hap1 422 415 7 1 2 99.2%
hap2 420 407 13 4 1 98.9%

Genome assembly stats for the verkko after polishing the genome with the ERR10930363.fastq reads

  • QUAST reports
haplotypes contigs Largest contig Total length GC (%) N50 N75 L50 L75
hap 1 823 37880684 534634347 35.47 15838572 9224508 12 23
hap 2 116 31333114 502764077 35.28 23599476 18852818 10 16

-BUSCO reports

assembly type complete busco complete and single copy busco complete and duplicated busco fragmented busco missing busco completion score
hap1 393 385 8 1 31 92.5%
hap2 394 383 11 3 28 92.7%
  • I am extending this with the HiC, Chromosomal Application development, Genome bubble solver and other applications.

Gaurav Sablok University of Potsdam Potsdam,Germany