University of Maryland Institute for Physical Science and Technology MaSurca 2 Genome Assembler Quick Start Guide
Yüklə 44,38 Kb.
|
- Bu səhifədəki naviqasiya:
- 1. System requirements/run rimes Compile/Install.
- Hardware requirements
- Expected run times
- 2. Installation instructions
- 3. Running the assembler Overview.
- Configuration
- The masurca and the assemble.sh script.
- Restarting a failed assembly.
- Assembly result.
|
University of Maryland Institute for Physical Science and Technology MaSuRCA-3.2.2 Genome Assembler Quick Start Guide The MaSuRCA (Maryland Super Read Cabog Assembler) assembler combines the benefits of deBruijn graph and Overlap-Layout-Consensus assembly approaches. Since version 3.2.1 it supports hybrid assembly with short Illumina reads and long high error PacBio/MinION data. Citation for MaSuRCA: imin AV, Marçais G, Puiu D, Roberts M, Salzberg SL, Yorke JA. The MaSuRCA genome assembler. Bioinformatics. 2013 Nov 1;29(21):2669-77. Citation for MaSuRCA hybrid assembler: Zimin AV, Puiu D, Luo MC, Zhu T, Koren S, Yorke JA, Dvorak J, Salzberg S. Hybrid assembly of the large and highly repetitive genome of Aegilops tauschii, a progenitor of bread wheat, with the mega-reads algorithm. Genome Research. 2017 Jan 1:066100. 1. System requirements/run rimes Compile/Install. To compile the assembler we require gcc version 4.7 or newer to be installed on the system. Only Linux is supported (May or may not compile under gcc for MacOS or Cygwin, Windows, etc). The assembler has been tested on the following distributions:
Hardware requirements. The hardware requirements vary with the size of the genome project. Both Intel and AMD x64 architectures are supported. The general guidelines for hardware configuration are as follows:
Expected run times. The expected run times depend on the cpu speed/number of cores used for the assembly and on the data used. The following lists the expected run times for the minimum configurations outlined above for Illumina-only data sets. Adding long reads (454, Sanger, etc. makes the assembly run about 50-100% longer:
2. Installation instructions To install, first download the latest distribution from ftp://ftp.genome.umd.edu/pub/MaSuRCA/. Then untar/unzip the package MaSuRCA-X.X.X.tgz, cd to the resulting folder and run './install.sh'. The installation script will configure and make all necessary packages. In the rest of this document, '/install_path' refers to a path to the directory in which './install.sh' was run. 3. Running the assembler Overview. The general steps to run the MaSuRCA assemblers are as follows, and will be covered in details in later sections. It is advised to create a new directory for each assembly project. IMPORTANT! Do not pre-process Illumina data before providing it to MaSuRCA. Do not do any trimming, cleaning or error correction. This WILL deteriorate the assembly. First, create a configuration file which contains the location of the compiled assembler, the location of the data and some parameters. Copy in your assembly directory the template configuration file '/install_path/sr_config_example.txt' which was created by the installer with the correct paths to the freshly compiled software and with reasonable parameters. Many assembly projects should only need to set the path to the input data. Second, run the 'masurca' script which will generate from the configuration file a shell script 'assemble.sh'. This last script is the main driver of the assembly. Finally, run the script 'assemble.sh' to assemble the data. Configuration. To run the assembler, one must first create a configuration file that specifies the location of the executables, data and assembly parameters for the assembler. The installation script will create a sample config file 'sr_config_example.txt'. Lines starting with a pound sign ('#') are comments and ignored. The sample configuration file looks like this: #example configuration file for rhodobacter sphaeroides assembly from GAGE project (http://gage.cbcb.umd.edu) #DATA is specified as type {PE,JUMP,OTHER}= two_letter_prefix mean stdev fastq(.gz)_fwd_reads fastq(.gz)_rev_reads #NOTE that PE reads are always assumed to be innies, i.e. ---> <---, and JUMP are assumed to be outties <--- --->; if there are any jump libraries that are innies, such as longjump, specify them as JUMP and specify NEGATIVE mean #IT IS MANDATORY to supply some Illumina paired end or single end reads #reverse (R2) reads are optional for PE libraries and mandatory for JUMP libraries #any OTHER sequence data (454, Sanger, Ion torrent, etc) must be first converted into Celera Assembler compatible .frg files (see http://wgs-assembler.sourceforge.com) and supplied as OTHER=file.frg; PACBIO reads must be in a single FASTA file and supplied as PACBIO=reads.fa; NANOPORE reads must be in a single FASTA file and supplied as NANOPORE=reads.fa DATA
PE= pe 180 20 /FULL_PATH/frag_1.fastq /FULL_PATH/frag_2.fastq JUMP= sh 3600 200 /FULL_PATH/short_1.fastq /FULL_PATH/short_2.fastq OTHER=/FULL_PATH/file.frg PACBIO=reads.fa END
#this is k-mer size for deBruijn graph values between 25 and 101 are supported, auto will compute the optimal size based on the read data and GC content. Do not set this longer than PE read length!!! GRAPH_KMER_SIZE=auto #set this to 1 for all Illumina-only assemblies #set this to 1 if you have less than 20x long reads (454, Sanger, Pacbio) and less than 50x CLONE coverage by Illumina, Sanger or 454 mate pairs #otherwise keep at 0 USE_LINKING_MATES=1 #this parameter is useful if you have too many jumping library mates. Typically set it to 60 for bacteria and something large (300) for mammals LIMIT_JUMP_COVERAGE = 60 #these are the additional parameters to Celera Assembler. do not worry about performance, number or processors or batch sizes -- these are computed automatically. for mammalian genomes do not set cgwErrorRate above 0.15!!! set cgwErrorRate=0.25 for bacteria CA_PARAMETERS = cgwErrorRate=0.15 # number of cpus to use NUM_THREADS= 64 #this is mandatory jellyfish hash size – 10x the genome size is a good starting value JF_SIZE=100000000 #this specifies if we do (1) or do not (0) want to trim long runs of homopolymers (e.g. GGGGGGGG) from 3' read ends, use it for high GC genomes DO_HOMOPOLYMER_TRIM=0 END
DATA – in this section the user must specify the types of data available for the assembly. Each line represent a library and must start with PE=, JUMP= or OTHER= for the 3 different type of input read library (Paired Ends, Jumping or other). There can be multiple lines starting with 'PE=' (or JUMP=), one line per library. PE and JUMP data must be in fastq format while the other data is in provided as a Celera Assembler frag format ('.frg'). Every PE or JUMP library is named by a unique two letter prefix. No two library can have the same prefix and a prefix should be made of two printable characters or number (no space or control characters), e.g. 'aa', 'ZZ', 'l5', or 'J2'. The following types of data are supported:
Here is an example of how the assembled contig N50 varies with Pacbio coverage on A. thaliana data set (from Zimin et al, 2017): More than one entry for each data type/set of files is allowed EXCEPT for PacBio/Nanopore data. That is if you have several pairs of PE fastq files, specify each pair on a separate line with a different two-letter prefix.
Optional parameters:
The masurca and the assemble.sh script. Once you’ve created a configuration file, use the ‘masurca' script from the MaSuRCA bin directory to generate the 'assemble.sh' shell script that executes the assembly: $ /install_path/ MaSuRCA-X.X.X/bin/masurca config.txt To run the assembly, execute 'assemble.sh'. Typically upon completion of the successful assembly, the current directory, where 'assemble.sh' was generated, will contain the following files, in reverse chronological order: $ ls –lth FILE CONTAINS gapClose.err STDOUT/STDERR for gap filling CA CABOG folder – the final assembly ends up in CA/9-terminator or CA/10-gapclose (if gapClose succeeded, most of the time) runCA2.out CABOG stdout for scaffolder tigStore.err Stderr for tigStore unitig_cov.txt CABOG coverage statistics global_arrival_rate.txt CABOG coverage statistics, global arrival rate unitig_layout.txt Unitig layout/sequences used to recomputed the coverage statistics genome.uid File relating UID to read name for CABOG runCA1.out CABOG stdout for unitig consensus runCA0.out CABOG stdout for initial stages: store building, overlapping, unitigging superReadSequences_shr.frg FRG file for super reads, super reads >2047br are shredded with an overlap of 1500bp renamed_sr.txt Deprecated file, ignore pe.linking.frg FRG file of PE pairs where the two reads ended up in different super reads pe.linking.fa Fasta file of PE pairs where the two reads ended up in different super reads work1 Working directory of the super reads code that generates the super reads from PE reads super1.err STDERR output of the super reads code that generates the super reads from PE reads sj.cor.clean.frg CABOG FRG file with corrected jumping library pairs, redundant and non-junction removed, coverage limited sj.cor.ext.reduced.fa Fasta file with corrected jumping library pairs, redundant and non-junction removed, coverage limited mates_to_break.txt File that lists the jumping library mates that are to be removed, if the jumping library clone coverage exceeds the LIMIT_JUMP_COVERAGE parameter compute_jump_coverage.txt Supplementary file used to compute clone coverage of the jumping library, post-filtering sj.cor.clean.fa Fasta file with corrected jumping library pairs, redundant and non-junction removed redundant_sj.txt Text file with names of the redundant jumping library mate pairs chimeric_sj.txt text file with names of the non-junction jumping library mate pairs work2 working directory for the super reads code used to filter the JUMP libraries for non-junction/redundant pairs work2.1 working directory for secondary jumping filter based on the variable k-mer size super2.err STDERR output of the super reads code used to filter the JUMP libraries for non-junction/redundant pairs guillaumeKUnitigsAtLeast32bases_all.fasta fasta file for k-unitigs (see the paper referred above) k_u_0 jellyfish hash created from all error corrected reads, and used to estimate the genome size ??.cor.fa error corrected JUMP reads, one such file for each library with '??' being the prefix. The ordering of the reads is arbitrary, but the pairs are guaranteed to appear together. No quality scores. error_correct.log log file for error correction pe.cor.fa error corrected PE reads. The ordering of the reads is arbitrary, but the pairs are guaranteed to appear together. No quality scores combined_0 special combined Jellyfish hash for error correction pe_data.tmp supplementary information to figure out the GC content and lengths of PE reads sj.renamed.fastq the ??.renamed.fastq file is created for each “JUMP= …” entry in the configuration file. These file(s) contain renamed reads in the fastq format meanAndStdevByPrefix.sj.txt auto-generated file of “fake” mean and stdev for non-junction jumping library pairs. The Illumina protocol states that these are about 200-700bp long pe.renamed.fastq the ??.renamed.fastq file is created for each “PE= …” entry in the configuration file. These file(s) contain renamed reads in the fastq format. meanAndStdevByPrefix.pe.txt auto-generated file of means and stdevs for PE reads assemble.sh the original assemble.sh script
For example
Assembly result. The final assembly files are under CA/10-gapclose and named 'genome.ctg.fasta' for the contig sequences and 'genome.scf.fasta' for the scaffold sequences. Yüklə 44,38 Kb. Dostları ilə paylaş:
|