Pseudodomonas syringae: Difference between revisions

From Cbcb
Jump to navigation Jump to search
Line 62: Line 62:


== Assemblies ==
== Assemblies ==
=== 454 ===
  /fs/szasmg2/Bacteria/Pseudomonas_syringae/Assembly/454/2007_1015_AMOSCmp-relaxed
  no trimming; MINCLUSTER=20 MAXTRIM=10 MAJORITY=50
  Stats:
  desc            #elem  min    max    mean    stdev  sum
  contigs        6131    43      8261    966.57  829.44  5926089
  pos_gaps        5622    1      10394  110.32  283.78  620259
  Slight improvement by doing alignment based trimming of the 454 reads


=== 454 + Solexa ===
=== 454 + Solexa ===

Revision as of 17:54, 20 February 2008

Pseudomonas syringae pv. tomato str. DC3000


Data

Originally sequenced and finished at TIGR: published Sept 2003

NCBI

 AA: no assembly
 TA 80,959 reads 
 Genome Project
 Taxonomy TaxId=223283

Chromosome + 2 plasmids:

 Name           Length    %GC    Info
 NC_004578.1    6,397,126 58.40  chromosome
 NC_004633.1    73,661    55.15  plasmid pDC3000A
 NC_004632.1    67,473    56.17  plasmid pDC3000B
 total          6,538,260
 Little similarity between the chromosome and plasmids.
 The 2 plasmids share a significant amount of DNA; see /fs/szasmg2/Bacteria/Pseudomonas_syringae/Data/nucmer/NC_004633-NC_004632.png

UNC: Jeff Dangl

New sequence:

Read stats

 Type   File                            #reads            min     median  max     sum             mean    stdev   n50
 Solexa DC3000.reads.filtered.fasta     6,340,136         32      32      32      202884352       32      0       32
 454p   DC3000.format.454Reads.fna      123,992           38      86      329     15623908        126.01  58.89   142
 454    DC3000.TCA.454reads.format.fna  77,466            35      244     371     18627363        240.46  26.85   245 
 * Solexa 3 lanes; 
 * 454 shotgun 1/4 Plate (250bp read); 
 * 454 paired ends 1/4 Plate : 
     * contain a 44 bp linker in the middle
     * the linker sequence is: GTTGGAACCGAAAGGGTTTGAATTCAAACCCTTTCGGTTCCAAC
     * there are some (not many) 454 paired end sequences that contain multiple instances of the linker (tandem): Example EUEIEUN01ANUGL_length=128_xy=0154_1891 
 
 
 Quality values are missing for all data sets!!!
 I assigned default qual=3 to all the base (.frg & .afg files)  

UNC sequence data: (not avail any more?)

 http://biology622.dhcp.unc.edu/~labweb/DCData/

UNC (e-mail):

 * Theoretical minimum number of contigs we can obtain is 268 (our reads fail to cover 269 nucleotides). 
 * Our de novo assembly spans the genome in 853 contigs totaling 6,313,026 bp. 
 * 98.7% of the genome is covered by a contig; 
 * 84% of the genome is covered by contigs 10,000 bp or greater. 
 * The average gap size between contigs is 98 bp; 
 * average contig size 7401 bp. 
 * The N50 = 37,444 bp. 
 * Our largest BAMBUS "scaffold" is 2,565,761 bp

Files location:

 /fs/szasmg2/Bacteria/Pseudodomonas_syringae/Data
 /fs/szasmg2/Bacteria/Pseudodomonas_syringae/Assembly

Assemblies

454

 /fs/szasmg2/Bacteria/Pseudomonas_syringae/Assembly/454/2007_1015_AMOSCmp-relaxed
 no trimming; MINCLUSTER=20 MAXTRIM=10 MAJORITY=50
 Stats:
 desc            #elem   min     max     mean    stdev   sum
 contigs         6131    43      8261    966.57  829.44  5926089
 pos_gaps        5622    1       10394   110.32  283.78  620259

 Slight improvement by doing alignment based trimming of the 454 reads

454 + Solexa

 /fs/szasmg2/Bacteria/Pseudodomonas_syringae/Assembly/Solexa-454/2007_1009_AMOSCmp-relaxed
 142 contigs (37 negative gaps, 89 positive gaps)
 No read trimming was done. 
 AMOScmp used the following parameters:
   nucmer -c  20
   casm-layout -t 20 -o 5
 "-t 20" allows for 20 bp long  dirty sequence ends which seem to solve the "low quality" problem.
 => 22 large contigs
 
 454 single reads + 30 bp Solexa reads  => 167 contigs , 49 negative gaps, 100 positive gaps 
 454 single reads + 25 bp Solexa reads  => 293 contigs,  144 negative gaps, 131 positive gaps 

454 + Solexa +454p

 Only the 454 paired ends that contain 1 single complete adaptor sequence were used (allmost all)
 /fs/szasmg2/Bacteria/Pseudodomonas_syringae/Assembly/Solexa-454-454p/2007_1011_AMOSCmp-relaxed-filtered
 149 contigs; very similar to the prev ome

454 + Solexa +454p

MAJORITY=50

 /fs/szasmg2/Bacteria/Pseudodomonas_syringae/Assembly/Solexa-454/2007_1015_AMOSCmp-relaxed-MAJORITY50 
 131 contigs  (18 negative gaps)
 No read trimming was done. 
 AMOScmp used the following parameters:
   nucmer -c  20
   casm-layout -t 20 -o 5 -m 50
 No read trimming was done. 
 "-t 20" allows for 20 bp long  dirty sequence ends which seem to solve the "low quality" problem.
 "-m 20" merges some contigs together
 => 10 large contigs
 contig#        len     gc%
 4              2290968 59.00
 7              1817904 58.18
 3              1405326 58.08
 5              648413  58.48
 2              192413  57.86
 6              87152   58.02
 131            71251   56.47
 1              32939   54.86
 130            29120   59.36
 9              20309   53.56
 95             3589    59.46


 Rerun Solexa32,Solexa30,Solexa25 with "nucmer -b 2 -g 5"
 
 2007_1015_AMOSCmp-relaxed-Solexa32/
 2007_1015_AMOSCmp-relaxed-Solexa30/
 2007_1015_AMOSCmp-relaxed-Solexa25/
 /fs/szasmg2/Bacteria/Pseudomonas_syringae/Assembly/Solexa-454/qc.combine.3
 
 $ show-coords 1con-contigs.delta | grep gi | awk '{print $7}' | getSummary.pl # sum of ref alignments: 13608985
 $ show-coords 1con-contigs.delta | grep gi | awk '{print $8}' | getSummary.pl # sum of qry alignments: 13747738
  138,753 bp in duplications for Solexa32 ??? 
   61,741 bp in duplications for Solexa30 ??? 
   10,881 bp in duplications for Solexa25 ??? 
 Copy of assembly files:
  /fs/ftp-cbcb/pub/data/dpuiu/Pseudomonas_syringae
  ftp://ftp.cbcb.umd.edu/pub/data/dpuiu/Pseudomonas_syringae/Solexa-454

Sanger AMOScmp

 Sanger reads
 /fs/szasmg2/Bacteria/Pseudodomonas_syringae/Assembly/Sanger/2007_1011_AMOSCmp-relaxed
 Many miss-oriented mates in the 4.8M-5M region of the chromosome
 22 contigs
 Chromosome
 Chromosome problem

Sanger Celera 3.11

 /fs/szasmg2/Bacteria/Pseudodomonas_syringae/Assembly/Sanger/2007_1011_WGA
 22 scaff, 46 contigs, 181 degens
 Scaffold 7180000001443 looks circular: possible 163,074 bp plasmid
 aligns to 4.8M-5M "problem" region in the chromosome
 7180000001443.png
     [S1]     [E1]  |     [S2]     [E2]  |  [LEN 1]  [LEN 2]  |  [% IDY]  |  [LEN R]  [LEN Q]  |  [COV R]  [COV Q]  | [TAGS]
 ===============================================================================================================================
        1   175592  |        1   175592  |   175592   175592  |   100.00  |   175592   175592  |   100.00   100.00  | 7180000001443   7180000001443   [IDENTITY]
        1    12519  |   163075   175592  |    12519    12518  |    99.98  |   175592   175592  |     7.13     7.13  | 7180000001443   7180000001443   [BEGIN]
   163075   175592  |        1    12519  |    12518    12519  |    99.98  |   175592   175592  |     7.13     7.13  | 7180000001443   7180000001443   [END]


     [S1]     [E1]  |     [S2]     [E2]  |  [LEN 1]  [LEN 2]  |  [% IDY]  |  [LEN R]  [LEN Q]  |  [COV R]  [COV Q] | [TAGS]
 ===============================================================================================================================
  4790727  4911492  |   120764        1  |   120766   120764  |    99.98  |  6397126   175592  |     1.89    68.78  | gi|28867243|ref|NC_004578.1|    7180000001443
  4898971  4955870  |   175592   118697  |    56900    56896  |    99.98  |  6397126   175592  |     0.89    32.40  | gi|28867243|ref|NC_004578.1|    7180000001443

Sanger AMOScmp (Chromosome+3 plasmids ref)

 Reference=complete genome(chromosome+3 plasmids) use "circular contig" in Celera 3.11 assembly
 /fs/szasmg2/Bacteria/Pseudodomonas_syringae/Assembly/Sanger/2007_1012_AMOSCmp-relaxed-3plasmids
 38 contigs: 15 for main chromosome, 1 for longer plasmid, 21 for shorter plasmid, 1 for "circular contig"
 The missoriented read pile corresponding to the chromosome (4. AMOSCmp of Sanger reads) has dissapeared
 AA ready for submission: /fs/szasmg2/Bacteria/Pseudodomonas_syringae/Assembly/Sanger/2007_1012_AMOSCmp-relaxed-3plasmids/AA/umd-20071030-141700.tar.gz

Solexa AMOScmp

1. align all reads (Solexa) to the reference using nucmer. I initially used minmatch=20, mincluster=20 (-c 20 -l 20)

 6340136 reads
 5641782 (88.98%) aligned by nucmer -c 20 -l 20
 3453618 (54.47%) aligned by nucmer -c 32 -l 20
 2707005 (42.69%) aligned by nucmer -c 32 -l 32


Solexa assemblied for different read coverages

Location

 /fs/szasmg2/Bacteria/Pseudomonas_syringae/Assembly/Solexa/sample/
 
 Several AMOScmp assemblies, using 10%,20%, ... 100%, of the P. syringae Solexa reads. 
 These would correspond to 3X,6X ... 30X, coverage 
 The read sampling was done randomly. One sample set for each coverage.

 all contigs
 desc    #elem   min     max     mean          stdev           sum
 10    43136   32      7712    135.11          140.61          5828148
 20    11243   32      20190   570.01          686.5           6408705
 30    2972    32      27962   2185.32         2804.56         6494784
 40    1058    32      63125   6152.98         7871.7          6509855
 50    455     32      163430  14319.01        19663.15        6515153
 60    267     32      328882  24406.61        46172.62        6516567
 70    166     32      671064  39260.9         84200.42        6517311
 80    143     32      906652  45577.16        111875.19       6517535
 90    117     32      1433643 55708.4         164246.61       6517883
 100   106     32      2067205 61489.83        230284.47       6517923
 
 chromo contigs
 desc    #elem   min     max     mean          stdev           sum
 10    42845   32      1845    133.32          118.36          5712348
 20    11124   32      9650    565.41          625.32          6289649
 30    2876    32      26076   2216.64         2714.92         6375063
 40    965     32      63125   6621.71         7893.19         6389957
 50    362     32      163430  17665.19        20565.31        6394800
 60    167     32      328882  38299.32        53660.75        6395987
 70    75      257     671064  85287.52        108858.19       6396564
 80    49      940     906652  130546.42       160470.1        6396775
 90    25      42603   1433643 255877.72       277650.54       6396943
 100   18      42603   2067205 355387.77       465907.88       6396980
 
 all gaps
 desc    #elem   min     max     mean    stdev   sum
 10    43137   1       3874    16.46   38.01   710112
 20    11242   1       3919    11.52   64.43   129555
 30    2971    1       3418    14.63   114.29  43476
 40    1056    1       3873    26.89   196.7   28405
 50    454     1       3415    50.89   291.04  23107
 60    265     1       3870    81.86   380.9   21693
 70    165     1       3868    126.96  486.88  20949
 80    141     1       3414    146.98  461.06  20725
 90    115     1       3418    177.19  520.11  20377
 100   104     1       3278    195.54  511.06  20337
 
 chromo gaps
 desc    #elem   min     max     mean    stdev   sum
 10    42846   1       240     15.98   16.33   684778
 20    11125   1       146     9.66    9.72    107477
 30    2876    1       76      7.67    7.73    22063
 40    965     1       58      7.42    7.8     7169
 50    362     1       48      6.42    7.08    2326
 60    167     1       58      6.82    7.63    1139
 70    76      1       55      7.39    7.9     562
 80    49      1       55      7.16    10.08   351
 90    25      1       45      7.31    10.12   183
 100   18      1       55      8.11    13.62   146

CBCB (new)

Alignment based trimming

!!! Reduced the duplications significantly

Solution:

1. align all reads (Solexa) to the reference using nucmer. I initially used minmatch=20, mincluster=20 (-c 20 -l 20)

 6340136 reads
 5641782 (88.98%) aligned by nucmer -c 20 -l 20
 3453618 (54.47%) aligned by nucmer -c 32 -l 20
 2707005 (42.69%) aligned by nucmer -c 32 -l 32