Abstract
The human genome is diploid, and knowledge of the variants on each chromosome is important for the interpretation of genomic information. Here we report the assembly of a haplotype-resolved diploid genome without using a reference genome. Our pipeline relies on fosmid pooling together with whole-genome shotgun strategies, based solely on next-generation sequencing and hierarchical assembly methods. We applied our sequencing method to the genome of an Asian individual and generated a 5.15-Gb assembled genome with a haplotype N50 of 484 kb. Our analysis identified previously undetected indels and 7.49 Mb of novel coding sequences that could not be aligned to the human reference genome, which include at least six predicted genes. This haplotype-resolved genome represents the most complete de novo human genome assembly to date. Application of our approach to identify individual haplotype differences should aid in translating genotypes to phenotypes for the development of personalized medicine.
This is a preview of subscription content, access via your institution
Access options
Subscribe to this journal
Receive 12 print issues and online access
$209.00 per year
only $17.42 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Accession codes
Primary accessions
Gene Expression Omnibus
Sequence Read Archive
Referenced accessions
Gene Expression Omnibus
Sequence Read Archive
References
Tewhey, R., Bansal, V., Torkamani, A., Topol, E.J. & Schork, N.J. The importance of phase information for human genomics. Nat. Rev. Genet. 12, 215–223 (2011).
Lander, E.S. et al. Initial sequencing and analysis of the human genome. Nature 409, 860–921 (2001).
Li, R. et al. SNP detection for massively parallel whole-genome resequencing. Genome Res. 19, 1124–1132 (2009).
Li, H. & Durbin, R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25, 1754–1760 (2009).
DePristo, M.A. et al. A framework for variation discovery and genotyping using next-generation DNA sequencing data. Nat. Genet. 43, 491–498 (2011).
Levy, S. et al. The diploid genome sequence of an individual human. PLoS Biol. 5, e254 (2007).
Wang, J. et al. The diploid genome sequence of an Asian individual. Nature 456, 60–65 (2008).
Ahn, S.-M. et al. The first Korean genome sequence and analysis: full genome sequencing for a socio-ethnic group. Genome Res. 19, 1622–1629 (2009).
Tong, P. et al. Sequencing and analysis of an Irish human genome. Genome Biol. 11, R91 (2010).
1000 Genomes Project Consortium. A map of human genome variation from population-scale sequencing. Nature 467, 1061–1073 (2010).
International HapMap Consortium. A haplotype map of the human genome. Nature 437, 1299–1320 (2005).
1000 Genomes Project Consortium. An integrated map of genetic variation from 1,092 human genomes. Nature 491, 56–65 (2012).
Browning, S.R. & Browning, B.L. Rapid and accurate haplotype phasing and missing-data inference for whole-genome association studies by use of localized haplotype clustering. Am. J. Hum. Genet. 81, 1084–1097 (2007).
Duitama, J. et al. Fosmid-based whole genome haplotyping of a HapMap trio child: evaluation of single individual haplotyping techniques. Nucleic Acids Res. 40, 2041–2053 (2012).
Woolley, A.T., Guillemette, C., Li Cheung, C., Housman, D.E. & Lieber, C.M. Direct haplotyping of kilobase-size DNA using carbon nanotube probes. Nat. Biotechnol. 18, 760–763 (2000).
Cao, H. et al. A short-read multiplex sequencing method for reliable, cost-effective and high-throughput genotyping in large-scale studies. Hum. Mutat. 34, 1715–1720 (2013).
Dapprich, J., Ferriola, D., Magira, E.E., Kunkel, M. & Monos, D. SNP-specific extraction of haplotype-resolved targeted genomic regions. Nucleic Acids Res. 36, e94 (2008).
Fan, H.C., Wang, J., Potanina, A. & Quake, S.R. Whole-genome molecular haplotyping of single cells. Nat. Biotechnol. 29, 51–57 (2011).
Kitzman, J.O. et al. Haplotype-resolved genome sequencing of a Gujarati Indian individual. Nat. Biotechnol. 29, 59–63 (2011).
Suk, E.-K. et al. A comprehensively molecular haplotype-resolved genome of a European individual. Genome Res. 21, 1672–1685 (2011).
Browning, S.R. & Browning, B.L. Haplotype phasing: existing methods and new developments. Nat. Rev. Genet. 12, 703–714 (2011).
Altshuler, D.M. et al. Integrating common and rare genetic variation in diverse human populations. Nature 467, 52–58 (2010).
Peters, B.A. et al. Accurate whole-genome sequencing and haplotyping from 10 to 20 human cells. Nature 487, 190–195 (2012).
Alkan, C., Coe, B.P. & Eichler, E.E. Genome structural variation discovery and genotyping. Nat. Rev. Genet. 12, 363–376 (2011).
Peng, Z. et al. Comprehensive analysis of RNA-seq data reveals extensive RNA editing in a human transcriptome. Nat. Biotechnol. 30, 253–260 (2012).
Li, Y. et al. The DNA methylome of human peripheral blood mononuclear cells. PLoS Biol. 8, e1000533 (2010).
Luo, R. et al. SOAPdenovo2: an empirically improved memory-efficient short-read de novo assembler. Gigascience 1, 18 (2012).
Zhang, G. et al. The oyster genome reveals stress adaptation and complexity of shell formation. Nature 490, 49–54 (2012).
Huang, W., Sherman, B.T. & Lempicki, R.A. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat. Protoc. 4, 44–57 (2009).
Wang, K., Li, M. & Hakonarson, H. ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic Acids Res. 38, e164 (2010).
Kirin, M. et al. Genomic runs of homozygosity record population history and consanguinity. PLoS ONE 5, e13996 (2010).
Schwartz, S. et al. Human-mouse alignments with BLASTZ. Genome Res. 13, 103–107 (2003).
Han, W. et al. Molecular cloning and characterization of chemokine-like factor 1 (CKLF1), a novel human cytokine with unique structure and potential chemotactic activity. Biochem. J. 357, 127–135 (2001).
Zook, J.M. et al. Integrating human sequence data sets provides a resource of benchmark SNP and indel genotype calls. Nat. Biotechnol. 32, 246–251 (2014).
Aghaei, M., Karami-Tehrani, F., Salami, S. & Atri, M. Adenosine deaminase activity in the serum and malignant tumors of breast cancer: the assessment of isoenzyme ADA1 and ADA2 activities. Clin. Biochem. 38, 887–891 (2005).
Li, R. et al. Building the sequence map of the human pan-genome. Nat. Biotechnol. 28, 57–63 (2010).
Trapnell, C. et al. Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat. Biotechnol. 28, 511–515 (2010).
Ng, P.C. SIFT: predicting amino acid changes that affect protein function. Nucleic Acids Res. 31, 3812–3814 (2003).
Adzhubei, I.A. et al. A method and server for predicting damaging missense mutations. Nat. Methods 7, 248–249 (2010).
Rusk, N. Causal mutations in a haploid landscape. Nat. Methods 9, 36 (2012).
Yang, M. & Park, J.Y. DNA methylation in promoter region as biomarkers in prostate cancer. Methods Mol. Biol. 863, 67–109 (2012).
Huang, W., Sherman, B.T. & Lempicki, R.a. Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res. 37, 1–13 (2009).
Adey, A., Burton, J., Kitzman, J. & Hiatt, J. The haplotype-resolved genome and epigenome of the aneuploid HeLa cancer cell line. Nature 500, 207–211 (2013).
Li, Y., Willer, C.J., Ding, J., Scheet, P. & Abecasis, G.R. MaCH: using sequence and genotype data to estimate haplotypes and unobserved genotypes. Genet. Epidemiol. 34, 816–834 (2010).
Khurana, E. et al. Integrative annotation of variants from 1092 humans: application to cancer genomics. Science 342, 1235587 (2013).
Pemberton, T.J. et al. Genomic patterns of homozygosity in worldwide human populations. Am. J. Hum. Genet. 91, 275–292 (2012).
Bentley, D.R. et al. Accurate whole human genome sequencing using reversible terminator chemistry. Nature 456, 53–59 (2008).
Li, Y. et al. Structural variation in two human genomes mapped at single-nucleotide resolution by whole genome de novo assembly. Nat. Biotechnol. 29, 723–730 (2011).
Lam, H.Y.K. et al. Performance comparison of whole-genome sequencing platforms. Nat. Biotechnol. 30, 78–82 (2012).
Audic, S. & Claverie, J.M. The significance of digital gene expression profiles. Genome Res. 7, 986–995 (1997).
Acknowledgements
We wish to recognize BGI-Shenzhen's sequencing platform for generating the data in this study. We thank the faculty and staff at BGI-Shenzhen who contributed to this project especially to H. Jiang and B. Wang, who helped on the experimental part, D. Cao, L. Lin and M. Li, who helped on the validation part, and L. Goodman, W. Chen and H. Jia, who provided critical comments on the manuscript. This work was supported by the State Key Development Program for Basic Research of China-973 Program (no. 2011CB809202); the Shenzhen Municipal Government of China (no. JC201005260191A); International Science and Technology Cooperation Program of Shenzhen (GJHZ20130417140835564); Tianjin Municipal Science and Technology special funds for Enterprise Development (no. 14ZXLJSY00320). G.W. was also supported by Alberta Innovates Technology Futures.
Author information
Authors and Affiliations
Contributions
Ju.W., G.W., Yi.L. and H.C. managed the project. Ju.W., G.K.-S.W., Yi.L., H.C., H.W., R.L. and S.H. designed the analyses. Ha.Y., construct the fosmid library. R.L., Y.S., Y.X., Bi.L., G.H., W.H., Bo.L., Z.H. and Yu.L. conducted assembly and evaluation; H.C., H.W., S.H., Y.S., X.T., H.Z., J.L., Y.W., F.Y., P.S., Si.L., P.G., H.H., J.S., D.C., L.C.A.M.T., X.L., Q.F., X.X., X.Z., L.B., K.K., R.D., S.D., R.N., So.L., Ji.W. and Hu.Y. performed the data analyses. H.C., H.W. and S.H. did most of the writing with contributions from all authors.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing financial interests.
Integrated supplementary information
Supplementary Figure 1 Number and length distributions of the inferred fosmids.
Fosmid were inferred by mapping each pool’s sequence reads to hg19. We retained only those reads that mapped to within a 3~50 kb region. (a) Fosmid number in each pool. On average, there were ~32 fosmids per pool. (b) Fosmid size. The average length was 36.8 kb.
Supplementary Figure 2 Fosmids physical coverage distribution.
Blue curve denotes the theoretical coverage distribution, at an average coverage of 8x, and red curve denote the actual coverage. The average fosmid coverage was 8x, with a median of 7x. About 7% of YHref was not covered by fosmids, which may be due to a bias in the fosmid library construction and/or sequencing.
Supplementary Figure 3 Completeness of assembled sequence in each fosmid pool.
The horizontal axis represents the percentage of the fosmid sequence that was assembled in each pool. The vertical axis represents the proportion of fosmid pools at that given percentage. In total, 88.5% of the assembled pools contained at least 80% of the fosmid sequence, and 53.2% of the assembled pools contained at least 95% of the fosmid sequence.
Supplementary Figure 4 Contiguity of assembled sequence for individual fosmids.
The horizontal axis represents the ratio of the longest assembled sequence vs the inferred length of each defined fosmid. The vertical axis represents the proportion of fosmids at the given ratio. 54.7% of fosmids had a longest assembled sequence equal to, or longer than, half of the fosmid length. About 18% of the fosmids were completely assembled.
Supplementary Figure 5 Construction of the haplotype-resolved sequence.
The top (orange) bar represents the non-phased YHref sequence and the bottom (multi-color) bar represents the haplotype-resolved output. The middle (blue) bars represent the fosmid assembled haploid (FAH) sequences belonging to the same haplotype.
Supplementary Figure 6 Theoretical N50 length of haplotype phasing and long homozygous region.
a. Long homozygous regions (>=20 kb) for different populations in 1000 genomes project. Asians have more long homozygous region than other populations. This might be why YH had a shorter haplotype N50 than other individuals sequenced at a comparable fosmid depth. b. The theoretical N50 length distribution of haplotype phasing using the method of the current study, in 4 different individuals. Heterozygous marker numbers are shown at the top-left. The haplotype N50 of YH is expected to be 510 kb with a fosmid coverage of 4x per haplotype (or 8x for a 3 Gb genome).
Supplementary Figure 7 HDG coverage on hg19 and RefSeq genes.
Our HDG sequence was aligned to the hg19 genome using Lastz. Coverage of the chromosomes and gene regions was calculated. “Both” means covered by the two assembled haplotypes (blue), “Single” means covered by just one assembled haplotype (red). a. Coverage information for each chromosome. b. Proportion of RefSeq gene at given coverage.
Supplementary Figure 8 Length distributions of insertions and deletions.
a. Length distribution of short indels (<10 bp). Peaks at multiples of 3 bp in the exon distribution are expected because they do not disturb the reading frames. b. Length distribution of long indels (100 bp~1 kb). The peak at ~300 bp is due an enrichment for Alu element insertions and deletions. Note that there is no bias between insertion and deletion, which is progress compared to previous studies. c. Distribution of long indels (100 bp~1 kb) in unique versus repeat regions. As expected, there are more indels in the repeat regions and the peak at ~300 bp is more pronounced. d. Length distribution of homozygous and heterozygous long indels (100 bp~1 kb).
Supplementary Figure 9 SNP detection and intersection from different methods/platforms.
A total of ~4.0 M SNPs were detected by three different methods/platforms. The majority (68.2%) of these was consistent between all three datasets. However, there were still tens of thousands of methods/platforms specific calls.
Supplementary Figure 10 Indel detection and intersection from different platforms.
We show the number of small indels detected by each method/platform and their intersection, at a flank size of 50 bp. For the ~1 M indels detected, there was only 27.6% concordance.
Supplementary Figure 11 Example of a heterozygous deletion located inside a gene.
This heterozygous deletion was detected by the ASV method but difficult to find by either WGS resequencing method. The yellow block in the reference is the region that was missing from hap2. Below are the WGS reads aligned to this region. This 151 bp deletion covered the 5-UTR and a part of exon1 for the gene PSMD1.
Supplementary Figure 12 Example of a heterozygous insertion located inside a gene.
This heterozygous insertion was detected by the ASV method but difficult to find by either WGS resequencing method. The yellow block in hap1 is the region that was missing from the reference. Below are the WGS reads aligned to this region. Near the breakpoint there were very few reads, perhaps because the insert sequence influenced the alignment. This 54 bp insertion covered exon3 of the gene LATS2.
Supplementary Figure 13 Variation rate for YH vs hg19 and heterozygosity between the two haplotypes of YH.
The curves at the top and the right summarize the distribution of heterozygosity rates for the two haplotypes of YH and the variation between YH and hg19, respectively. The black line indicates the 99% cutoff for each distribution.
Supplementary Figure 14 The classification of novel gene sequences.
a. Classification of different types of novel and gap covered sequences. i) novel insertion; ii-iv), novel haplotypes; v-vii), gap covered sequences; viii), orphan scaffolds. b. Distribution of novel sequences based on their length and number, in 100 bp bins. Novel sequences of length >1000 bp accounted for 93% of the total length. The longest was 123 kb. c. Distribution of breakpoints for novel sequences. Most of the novel sequences were in non-coding (intron, repeat and intergenic) regions. Only 0.8% were in CDS regions. These distributions are subdivided by the length of the sequence, represented by the color bars. d. Repeat content based on RepeatMasker.
Supplementary Figure 15 Examples of cis- and trans-acting genes.
a. Cis-acting gene DSPP on 4q22.1 encoding dentin sialophosphoprotein. Mutations in DSPP are associated with Dentinogenesis imperfecta, Shields type II, and deafness. b. Trans-acting gene CA9 on 9p13.3. Diseases associated with mutations in CA9 include horseshoe kidney and renal cell carcinoma. GO annotations include carbonate dehydratase activity.
Supplementary Figure 16 Allele specific methylation and expression.
Venn diagram showing the relationship between allele specific methylation (ASM) and allele specific expression (ASE). The numbers refer to the gene count. The red/brown circle inside the larger ASM circle represents genes where ASM was detecting in the promoter region.
Supplementary Figure 17 Construction of the fosmid libraries.
Approximately 30 fosmid clones were cultured together to form a single fosmid pool. Then, 3 μg of DNA from each pool was digested, and fragments with insert size ranging from 180 to 800 bp were selected. Adapters containing the 11 bp barcode were ligated to these selected fragments to form a single pooled-fosmid library. Barcoded fragments from 60~320 single pooled-fosmid libraries were pooled again (evenly) to create a Stage I barcode library. DNA fragments of sizes between 180 bp to 650 bp (lengths exclude barcode) from each Stage I barcode library were used to construct two independent libraries (one with small insert sizes and one with intermediate insert sizes). Each library was then PCR amplified with index primers, each of which contained an 8 bp barcode, to form a Stage II barcode library.
Supplementary Figure 18 Indel positional concordance as a function of flank size for the different methods of detection.
To determine the best flank size for use in indel detection, we plotted the concordance between the ASV and resequencing based analyses. The results stabilize at above 50 bp.
Supplementary Figure 19 Length distributions for method-specific short indels.
Short indels (1-50 bp) detected only by one method/platform, were selected out and plotted according to the length. Top-right figure provided information for indels with length between 10 and 50.
Supplementary Figure 20 Example of ASV-specific indel supported by fosmid aligned reads.
This was a 3 bp heterozygous deletion in a region covered by fosmids from eight independent pools, two of which supported the deletion.
Supplementary information
Supplementary Text and Figures
Supplementary Figures 1–20 and Supplementary Tables 1–3 and 5 (PDF 1737 kb)
Supplementary Table 2
Genes in hypervariable regions (XLSX 49 kb)
Supplementary Table 3
Annotation of predicted novel genes (XLSX 12 kb)
Supplementary Table 4
Cis and Trans genes annotation (XLSX 21 kb)
Supplementary Table 5
ASE and ASM gene analysis (XLSX 247 kb)
Rights and permissions
About this article
Cite this article
Cao, H., Wu, H., Luo, R. et al. De novo assembly of a haplotype-resolved human genome. Nat Biotechnol 33, 617–622 (2015). https://doi.org/10.1038/nbt.3200
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/nbt.3200
This article is cited by
-
A haplotype-resolved genome assembly of the Nile rat facilitates exploration of the genetic basis of diabetes
BMC Biology (2022)
-
Haplotype-resolved genomes provide insights into structural variation and gene content in Angus and Brahman cattle
Nature Communications (2020)
-
Chromosome-level assembly of the water buffalo genome surpasses human and goat genomes in sequence contiguity
Nature Communications (2019)
-
Platanus-allee is a de novo haplotype assembler enabling a comprehensive access to divergent heterozygous regions
Nature Communications (2019)
-
Better ILP models for haplotype assembly
BMC Bioinformatics (2018)
