Genome Sequencing and Assembly High throughput Sequencing

1
Genome Sequencing and AssemblyHigh throughput
Sequencing

• Xiaole Shirley Liu
• Jun Liu
• STAT115, STAT215

2
Outline

• Genome sequencing strategy
• Clone-by-clone sequencing
• Whole-genome shotgun sequencing
• Hybrid method
• Next generation sequencing technologies

3
Competing Sequencing Strategies

• Clone-by-clone and whole-genome shotgun

4
Clone-by-Clone Shotgun Sequencing

• E.g. Human genome project
• Map construction
• Clone selection
• Subclone library construction
• Random shotgun phase
• Directed finishing phase and sequence
  authentication

5
Map Construction

• Clone genomic DNA in YACs (1MB) or BACs (200KB)
• Map the relative location of clones
• Sequenced-tagged sites (STS, e.g. EST) mapping
• PCR or probe hybridization to screen STS
• Restriction site fingerprint
• Most time consuming
• 1990-98 to generate physical maps for human
  http//www.ncbi.nlm.nih.gov/genemap99/

6
Resolve Clone Relative Location

• Find a column permutation in the binary
  hybridization matrix, all ones each row are
  located in a block

STS 1 2
3 4 5
DNA clone 1 clone 2 clone 3 clone
4
3 5 1 4 2
1 1 0 0 1 0
2 0 0 1 0 1
3 1 0 0 1 1
4 1 1 0 1 0
1 2 3 4 5
1 0 0 1 1 0
2 1 1 0 0 0
3 0 1 1 1 0
4 0 0 1 1 1
STS Clone
7
Clone Selection

• Based on clone map, select authentic clones to
  generate a minimum tiling path
• Most important criteria authentic

8
Subclone Library Construction

• DNA fragmented by sonication or RE cut
• Fragment size 2-5 KB

9
Random Shotgun Phase

• Dideoxy termination reaction
• Informatics programs
• Coverage and contigs

10
Dideoxy Termination

• Method invented by Fred Sanger
• Automated sequencing developed by Leroy Hood
  (Caltech) and Michael Hunkapiller (ABI)

11
Bioinformatics Programs

• Developed at Univ. Wash
• Phred
• Base calling
• Phil Green
• Phrap
• Assembly
• Brent Ewing
• Consed
• Viewing and editing
• David Gordon

12
Coverage and contigs

• Coverage sequenced bp / fragment size
• E.g. 200KB BAC, sequenced 1000 x 500bp subclones,
  coverage 1000 x 500bp / 200KB 2.5X
• Lander-Waterman curve

13
Directed Finishing Phase

• David Gordon auto-finish
• Deign primers at gap 2 ends, PCR amplify, and
  sequence the two ends until they meet
• Sequence authentication verify STS and RE sites
• Finished lt 1 error (or ambiguity) in 10,000bp,
  in the right order and orientation along a
  chromosome, almost no gaps.

14
Genome-Shotgun Sequencing

• Celera human and drosophila genomes
• No physical map
• Jigsaw puzzle assembly
• Coverage 7-10X

15
Shotgun Assembly

• Screener
• Identify low quality reads, contamination, and
  repeats
• Overlapper
• gt 40bp overlap with lt 6 mismatches
• Unitigger
• Combine the easy (unique assembly) subset first
• Scaffolder repeat resolution
• Generate different sized-clone libraries, and
  just sequence the clone ends (read pairs)
• Use physical map information if available
• Consensus

16
Hybrid Method
17
Hybrid Method

• Optimal mixture of clone-by-clone vs whole-genome
  shotgun not established
• Still need 8-10X overall coverage
• Bacteria genomes can be sequenced WGS alone
• Higher eukaryotes need more clone-by-clone
• Comparative genomics can reduce the physical
  mapping (clone-by-clone) need
• Sequencing cost decreasing quickly
• Goal 1000 / genome

18
First Generation Sequencing
19
Second Generation Sequencing
20
2nd Gen Sequencing Tech

• Traditional sequencing machine
• 384 reads 1kb / 3 hours
• 454 (Roche)
• 1M reads 400bp / 5 hours
• Solexa (Illumina)
• 100M-1B reads of 30-100bp / 3-8 days, 8-16
  samples
• SOLiD (Applied Biosystems)
• 1.4B reads of 35-50bp / 5-8 days, 16 samples
• Helicos (single molecule sequencing)
• 500M reads of 30 bp / week, 50 samples
• Moving targets

21
Illumina (Solexa) Workflow
22
Illumina HiSeq2000

• Throughput
• 1100 / lane
• 35-100 bp / read
• 16 lanes (2 flow cells) / run
• 60-80 million reads / lane
• Sequencing a human genome 10000, 1 week
• Bioinfo challenges
• Very large files
• CPU and RAM hungry
• Sequence quality filtering
• Mapping and downstream analysis

23
Seq Files
_at_HWI-EAS3051119910/1 GCTGGAGGTTCAGGCTGGCCGGAT
TTAAACGTAT HWI-EAS3051119910/1 MVXUWVRKTWWUL
RQQMMWWBBBBBBBBBBBBBB _at_HWI-EAS3051112010/1 AA
GACAAAGATGTGCTTTCTAAATCTGCACTAAT HWI-EAS305111
2010/1 PXXXTXYXTTWYYYXXWWWTMTVXWBBB HWUSI
-EAS366_0112611298188280/1    16      chr9  
 98116600        255     38M           0      
0       TACAATATGTCTTTATTTGAGATATGGATTTTAGGCCG
 Y\bcdab\_UULbTUT\ccLbbYaYcWLYW  XAi1
 MDZ3C30T3     NMi2 HWUSI-EAS366_0112611257
188190/1    4             0       0        
          0       0       AGACCACATGAAGCTCAAGAAG
AAGGAAGACAAAAGTG  ecedddT\cTcaccdK\c__Yb\_c
KS_W\  XMi1 HWUSI-EAS366_0112611315195290/
1    16      chr9    102610263       255     38M
          0       0       GCACTCAAGGGTACAGGAAAAG
GGTCAGAAGTGTGGCC  c_Yc\LcbbbYdTa\dd\ddacdd\Y\d
ddcT  XAi0  MDZ38 NMi0 chr1 123450 123500
chr5 28374615 28374615 -

• Raw FASTQ
• Sequence ID, sequence
• Quality ID, quality score
• Mapped SAM
• Map 0 OK, 4 unmapped, 16 mapped reverse strand
• MD mismatch info
• NM number of mismatch
• Mapped BED
• Chr, start, end, strand

24
(Potential) Applications

• Metagenomics and infectious disease
• Ancient DNA, recreate extinct species
• Comparative genomics (between species) and
  personal genomes (within species)
• Genetic tests and forensics
• Circulating nucleic acids
• Risk, diagnosis, and prognosis prediction
• Transcriptome and transcriptional regulation
• More later in the semester

25
Third Generation Sequencing

• Single molecule sequencing (no amplification
  needed)
• Some can read very long sequences
• In 2-3 years, the cost of sequencing a human
  genome will drop below 1000
• Personal genome sequencing will be a key
  component of public health in every developed
  country
• The cost of sequencing will be lower than the
  cost of storing the sequences
• Bioinformatics will be key to convert data into
  knowledge

26
HW Q for Graduate Students

• Write the first page (Specific Aims) of an NIH
  research proposal with 1 million budget that
  uses high throughput sequencing and
  bioinformatics analysis to solve some interesting
  biomedical problems

27
How to Write a Specific Aims Page

• Grant title and your name
• Introductory (1-2) paragraphs (1/4-1/3 page)
• A is very important in bio/medicine/disease
• Recent development in A has made some really
  significant findings or improvements
• However, something is still lacking or not known
  about A
• The long term goal of our research is
• The focus of this proposal is
• The central hypothesis of this proposal is
• Therefore, we plan to do investigate / develop

28
How to Write a Specific Aims Page

• Specific Aims (1/3 to ½ page)
• Specifically, we plan to
• Aim 1 profile the genome-wide xxx of xxx
• 1.1 establish xxx
• 1.2
• Aim 2 develop a computer algorithm or knowledge
  base
• 2.1 model xxx
• Aim 3 identify the mechanism of xxx

29
Specific Aims

• Sound like you can definitely do it
• Do not use words that sound like a fishing
  expedition such as try, explore (find) or words
  that sound too trivial such as download (collect)
• Try not to let your aims successes depend on
  each other, but it is also important to be
  intellectually coherent among the aims (not a
  collection of unrelated topics)
• Propose as many aims as years for the grant

30
How to Write a Specific Aims Page

• Last paragraph (lt ¼ page)
• Whats novel about our approach
• Deliverables (what will the scientific community
  see at the end)
• A software, database, map, mechanism, resource
• Whats the (potential) significance about our
  proposal
• Best possible outcome

31
Summary

• Genome sequencing and assembly
• Clone-by-clone HGP
• Map big clones, find path, shotgun sequence
  subclones, assemble and finish
• Sequencing dideoxy termination
• Whole genome shotgun Celera
• Massively Parallel Sequencing
• 454, Solexa, SOLiD, Helicos
• Many opportunities and many challenges
• Project proposal

32
Acknowledgement

• Fritz Roth
• Dannie Durand
• Larry Hunter
• Richard Davis
• Wei Li
• Jarek Meller
• Stefan Bekiranov
• Stuart M. Brown
• Rob Mitra