Genome closure and finishing

1
Genome closure and finishing
2
Clone walking
clone insert (from shotgun library)
primer locations
Already sequenced
Unsequenced
sequencing primer
direction of extension
tgcatgatcgtgatcat acgtactagcactagtactgtagtcgatgcac
tgatcgatcgatcgatgctacgatgcatgct...
3
PCR to close gaps
Scaffold
Design primers at these locations

• Run PCR for each pair of primers
• Primers must be in non-repeat sequences
• Primers must be as close as possible to each gap
• PCR works best if primers are less than 2kb apart

4
Figure 1 Finishing procedures with Autofinish and
with a human finisher
David Gordon et al. Genome Res. 2001 11 614-625
5
Figure 2 Gap Closing Autofinish-Hybrid vs.
Human-only for five different BACs
David Gordon et al. Genome Res. 2001 11 614-625
6
Figure 3 Finishing reads required
Autofinish-Hybrid vs. Human-only
David Gordon et al. Genome Res. 2001 11 614-625
7
What about physical gaps?
How to order and orient all scaffolds?
8
(No Transcript)
9
Traditional strategy combinatorial PCR

• Design primers for each end of each scaffold
• With N/2 scaffolds and N ends, run a separate
  PCR reaction for every pair of ends
• (N choose 2) reactions N(N-1)/2
• Example 24 physical gaps, 48 ends 1128 PCR
  experiments

10
Multiplex PCR
actgagatatac
gttgagatataa
gcgacgctgctc
ccagcgctgttc
11
Multiplexing more primers

• Up to 12 primers per tube
• If any two primers surround the same gap, they
  produce a product
• If more than 2 pairs react, we get multiple
  products
• Let N number of primers 48
• K max primers/tube 12

12
POMPpipette optimized multiplex PCR

• minimize number of laboratory reactions
• with N48, number of combinatorial PCRs is 1128
• number of pipettings 2256
• POMP 28 reactions, 104 pipettings

13
POMP

• Create pools of size K/2 6
• Each of N48 primers put into 1 pool
• So we create N/(K/2) 8 pools, P1...P8
• Now create multiplex PCR reactions with all pairs
  of pools
• (8 choose 2) 28 reactions needed

14
POMP

• Guarantees that all primers are tested against
  all others
• Each reaction tests (12 choose 2) 66 pairs
• Protocol tests 6628 1848 pairs
• Only 1128 distinct pairs, so POMP has some
  redundancy - some pairs appear in more than one
  reaction

15
How many pipette operations?

• 48 pipettings to create the pools (one for each
  primer)
• 2 pipettings to create each multiplex reaction
  mix
• Total 48 2(28) 104

16
Results Streptococcus pneumoniae

• N48 primers, N/2 24 scaffolds, 24 gaps
• 19 products observed in the first experiment
• Q how many gaps closed?

17
Interpreting results

• Case 1 product appears with Pi and Pj, but not
  in any other lane containing Pi
• see P2P6 or P7P8 on previous slide
• purify and sequence product directly
• Case 2 product appears in Pi and Pj and also in
  other lanes containing Pi
• thus two primers within Pi reacted
• see pool P5 on previous slide

18
A Deconvoluting pool P5 eliminate each of the
six primers from the pool and run 6 standard PCRs
Answer p25 and p29
19
B Pools P2 and P3 gave two products

• Could run 12 more PCRs eliminate each of 12
  primers and re-run multiplex PCR
• However, 5 of the 12 were eliminated by other
  results
• For example, primer p10 was eliminated by results
  from P1-P2

• Answers
• p11 in Pool 2 pairs with p13 in Pool 3
• p16 in Pool 3 pairs with p8 in Pool 2

20
POMP requirements

• (Ideally) If no restriction on K, choose K based
  on N (where N 2 x (gaps))
• Make pools of size K/2

Reactions
Pipettings
21
POMP summary for S. pneumoniae

• Out of 48 gaps, 42 were closed
• Strategy employs slightly under N/2 reactions,
  thus expected number of products per tube is 1
• This was borne out in experiments
• This became a standard lab technique at TIGR

22
(No Transcript)