de novo assembly - oregon state universitypeople.oregonstate.edu/~knausb/rna_seq/10_denovo.pdfbasic...

De novo assembly Brian J. Knaus

USDA Forest Service

Pacific Northwest Research Station

Kelly Vining and Katherine Pillman

Oregon State University

1

Calendar

December meeting cancelled.

January meeting organizational.

◦ Introductions.

◦ Projects we’re involved with.

◦ What we want to get out of this group?

February meeting: Cufflinks (pending)

2

Overview

What are de Bruijn graphs?

Software:

◦ Velvet

◦ ABySS

◦ Trinity

3

Genomic vs. transcriptomic

Genomic expectations:

◦ Long contigs (i.e., chromosomes).

◦ Uniform coverage.

Transcriptomic expectations:

◦ Short contigs (i.e., genes or operons).

◦ Non-uniform coverage.

4

Two types of computer error

Errors which crash the program and

generate an error message.

Errors where the software appears to

exit successfully but has actually done

something bad.

5

Essential bioinformatic software Database/query applications

◦ Blast

◦ Blat

◦ MAQ

◦ Bowtie

◦ BWA

De novo assembly

◦ Velvet

◦ ABySS

Soap 6

De novo assembly

Compiled code (usually C; Dennis Ritchie

R.I.P).

Require LOTS of memory!

Implement de Bruijn graphs.

7

8

ACTGATTG

9

ACTGATTG

k-mer = 5

10

ACTGATTG

k-mer = 5

11

ACTGATTG

k-mer = 5

12

ACTGATTG

k-mer = 5

13

ACTGATTG

k-mer = 5

Schematic representation of our implementation of the de Bruijn graph.

Zerbino D R , Birney E Genome Res. 2008;18:821-829

©2008 by Cold Spring Harbor Laboratory Press

Example of Tour Bus error correction.



Example of Tour Bus error correction.



Issues:

• Bubbles.

• Tips.

• Errors.

Velvet

Kelly Vining


Forestry

18

Velvet • Current version: 1.1.06 (1.1.05 at CGRB cluster)

• Zerbino DR, Birney E (2008) Velvet: algorithms for de novo short read assembly using de Bruijn graphs. Genome Res 18: 821–829.

• Zerbino DR, McEwen GK, Margulies EH, Birney E, 2009 Pebble and Rock Band: Heuristic Resolution of Repeats and Scaffolding in the Velvet Short-Read de Novo Assembler. PLoS ONE 4(12): e8407. doi:10.1371/journal.pone.0008407

Basic Velvet de novo assembly

Run velveth Velveth takes in a number of sequence files, produces a hashtable, then

outputs two files in an output directory (creating it if necessary),

Sequences and Roadmaps, which are necessary to velvetg.

Run velvetg Velvetg is the core of Velvet where the de Bruijn graph is built then

manipulated.

Note that although velvetg saves some files during the process to avoid

useless recalculations, the parameters are not saved from one run to the

next.

On the command line…

/local/cluster/velvet

[viningk@needleman velvet]$ ll

total 32

drwxr-xr-x 2 root root 34 Sep 24 2007 VELVET/

drwxr-xr-x 8 root root 4096 Jul 7 2009 velvet_0.7.42/

drwxr-xr-x 8 root root 4096 Dec 23 2009 velvet_0.7.55/

drwxr-xr-x 8 root root 4096 Feb 16 2010 velvet_0.7.58/

drwxr-sr-x 8 4205 1304 4096 Oct 4 2010 velvet_1.0.12/

drwxr-xr-x 8 root root 4096 May 18 10:29 velvet_1.1.02/

drwxr-xr-x 8 boyda spatafojlab 4096 Sep 7 14:39 velvet_1.1.05/

lrwxrwxrwx 1 root root 13 Sep 22 14:15 velvet -> velvet_1.1.05/

drwxr-xr-x 9 root root 152 Sep 22 14:15 ./

drwxrwxr-x 123 root cgrb 4096 Nov 1 13:34 ../

/local/cluster/velvet/velvet_1.1.05

[viningk@needleman velvet_1.1.05]$ ll

total 884

drwxr-xr-x 3 boyda spatafojlab 23 Jul 1 2009 third-party/

drwxr-xr-x 3 boyda spatafojlab 23 Dec 9 2009 doc/

drwxr-xr-x 2 boyda spatafojlab 90 Dec 21 2010 data/

-rwxr-xr-x 1 boyda spatafojlab 26 Jan 11 2011 update_velvet.sh*

-rwxr-xr-x 1 boyda spatafojlab 479 Jan 11 2011 shuffleSequences_fastq.pl*

-rwxr-xr-x 1 boyda spatafojlab 876 Jan 11 2011 shuffleSequences_fasta.pl*

-rw-r--r-- 1 boyda spatafojlab 17988 Jan 11 2011 LICENSE.txt

-rw-r--r-- 1 boyda spatafojlab 1117 Mar 8 2011 README.txt

-rw-r--r-- 1 boyda spatafojlab 456 Mar 8 2011 For_MAC_or_SPARC_users.txt

-rw-r--r-- 1 boyda spatafojlab 1956 Mar 29 2011 CREDITS.txt

-rw-r--r-- 1 boyda spatafojlab 3834 Mar 29 2011 globals.h

drwxr-xr-x 15 boyda spatafojlab 4096 Mar 29 2011 contrib/

drwxr-xr-x 2 boyda spatafojlab 4096 Jul 27 14:35 src/

-rw-r--r-- 1 boyda spatafojlab 6118 Jul 28 13:38 Makefile

-rw-r--r-- 1 boyda spatafojlab 10871 Jul 28 13:38 ChangeLog

drwxr-xr-x 3 boyda spatafojlab 4096 Sep 7 14:39 obj/

-rwxr-xr-x 1 boyda spatafojlab 64819 Sep 7 14:39 velveth*

-rwxr-xr-x 1 boyda spatafojlab 246830 Sep 7 14:39 velvetg*

-rw-r--r-- 1 boyda spatafojlab 439712 Sep 7 14:39 Manual.pdf

-rw-r--r-- 1 boyda spatafojlab 61416 Sep 7 14:39 Columbus_manual.pdf

drwxr-xr-x 8 boyda spatafojlab 4096 Sep 7 14:39 ./

drwxr-xr-x 9 root root 152 Sep 22 14:15 ../

Velveth Designate file type (fastq, fasta, etc)

Designate read category

Designate hash size

Supported file formats are:

fasta (default)

fastq

fasta.gz

fastq.gz

sam

bam

eland

gerald

Read categories are:

short (default)

shortPaired**

short2 (same as short, but for a separate

insert-size library)

shortPaired2** (see above)

long (for Sanger, 454 or even reference

sequences)

longPaired**

**Paired-end reads must be in a single file (ShuffleSequences.pl)

Sample commands: velveth

> ./velveth output_directory hash_length

[[-file_format][-read_type] filename]

> /deepspace/strauss/viningk/velvet_1.1.02/velveth velvet_041411 29

-fastq -shortPaired s_5_seqsTrim6.txt

> /local/cluster/velvet/velvet_1.1.02/velveth velvet_091410b 31

-fastq -short s_1_seqsMergedTrim6.txt -fasta -long contigs_09.fa

Velvetg Designate library insert size

Designate minimum contig length

Designate coverage cutoff

Need to optimize according to genome size, expected genome coverage depth,

etc.

Examples: > ./velvetg output_directory/ -min_contig_lgth 100

> /deepspace/strauss/viningk/velvet_1.1.02//velvetg velvet_041411/

-ins_length 350 -min_contig_lgth 500

> /deepspace/strauss/viningk/velvet_1.1.02/velvetg velvet_051811b

-ins_length 350 -exp_cov 20 -cov_cutoff 7.9 -min_contig_lgth 500

Velvet Optimiser

Tries a range of k-mer lengths with velveth, then runs velvetg

with range of coverage cutoffs

Lives here on the CGRB cluster:

/local/cluster/velvet/velvet_1.1.05/contrib/VelvetOptimiser-2.1.7

Sample command:

/deepspace/strauss/viningk/velvet_1.1.02/contrib/VelvetOptimiser-

2.1.7/VelvetOptimiser.pl -s 27 -e 31 -f '-shortPaired -fastq s_5_seqsTrim6.txt

s_1_seqsMergedTrim6.txt' > s_15_VOtest3.txt

Sample VelvetOptimiser output ********************************************************

Assembly id: 1

Velveth timestamp: May 20 2011 13:47:08

Velveth version: 1.0.12

Readfile(s): -shortPaired -fastq s_5_seqsTrim6.txt

s_1_seqsMergedTrim6.txt

Velveth parameter string: auto_data_31 31 -shortPaired -fastq

s_5_seqsTrim6.txt s_1_seqsMergedTrim6.txt

Assembly directory: /deepspace/strauss/viningk/auto_data_31

Velvet hash value: 31

Roadmap file size: 12143769440

Total number of sequences: 150746258

**********************************************************

May 20 13:47:08

Beginning vanilla velvetg runs.

********************************************************

Assembly id: 1

Assembly score: 140


Velvetg timestamp: May 20 2011 18:59:34


Velvetg version: 1.0.12





Velvetg parameter string: auto_data_31





Total number of contigs: 2635009

n50: 140

length of longest contig: 6491

Total bases in contigs: 343232524

Number of contigs > 1k: 11471

Total bases in contigs > 1k: 15400103

Sample VelvetOptimiser output

(continued) **********************************************************

May 21 06:11:49 Setting cov_cutoff to 5.195.

********************************************************

Assembly id: 1

Assembly score: 93102316


Velvetg timestamp: May 21 2011 18:22:55


Velvetg version: 1.0.12





Velvetg parameter string: auto_data_31 -exp_cov 17 -

cov_cutoff 5.1952





Total number of contigs: 1080880

n50: 602

length of longest contig: 12710

Total bases in contigs: 261325501

Number of contigs > 1k: 52009

Total bases in contigs > 1k: 93102316

Paired Library insert stats:

Paired-end library 1 has length: 154, sample standard deviation:

57

Paired-end library 1 has length: 160, sample standard deviation:

63

**********************************************************

ABySS

Katherine Pillman


Horticulture

http://www.molecularevolution.org/molevol

files/presentations/comparative-genomics-

abyss-20110714.pdf

29

http://www.molecularevolution.org/molevolfiles/presentations/comparative-genomics-abyss-20110714.pdf








de novo assembly - oregon state universitypeople.oregonstate.edu/~knausb/rna_seq/10_denovo.pdfbasic...

Documents