PAKAP

INTRODUCTION (from paper?)

We have established a Present/Absent Kmer Analysis Pipeline (PAKAP). By reducing each read to component k-mers and comparing the relative abundance of these sub-sequences, we overcome statistical limitations of whole read comparative analysis.

PAKAP consists of a series of scripts written in Perl, Python and Bash scripts and requires Jellyfish [Marcais 2011] as well as optionally SOAPaligner. The scripts are freely available for non-commercial use.

What does PAKAP depend on?

• Jellyfish for fast kmer counting
• Some non-standard Perl modules:
  • bioperl
    • Bio::SeqIO
    • Bio::SearchIO
  • Parallel::ForkManager
  • Statistics::Descriptive
  • Config::IniFiles
  • GD::Graph::linespoints (for the script identifyKmerSize)
• Optional: SOAPaligner

Download

• Latest Version 1.0:
  • INSERT LINK

How to install?

• Download the [link_here DiffKAP package].

How to run?

• Create your project configuration file by using the example config file. Here it is:

<nowiki>

[PIPELINE]

1. This should be the path to the directory containing the pipeline scripts.

SCRIPTS_PATH = /ebi/bscratch/jarrah/uqphilippb/2014_4_17_FixPAV/CSR_pipeline

1. OPTIONAL: Path to reference

REFERENCE_PATH = /ebi/bscratch/jarrah/uqphilippb/2014_4_17_FixPAV/CSR_pipeline/A.thaliana.whole.fasta

1. OPTIONAL: Path to gene annotation on reference

GFF_PATH =

[JELLYFISH]

1. The path to the local installation of jellyfish (http://www.cbcb.umd.edu/software/jellyfish/)

PATH = /work2/NCISF/MAS/W26/jarrah/apps/bin

1. The minimum k-mer size will be checked when KMER_SIZE=AUTO. Default is 5

MIN_KMER_SIZE=5

1. The maximum k-mer size will be checked when KMER_SIZE=AUTO, jellyfish limits to the maximum k-mer size to 31. Default is 22

MAX_KMER_SIZE=22

[IDENTIFYKMERSIZE]

1. Number of CPU will be used. Be aware of that the total amount of memory will be shared among all CPUs.

NUM_OF_PROCESSOR=4

1. treatment 1 ID which will be used for naming files

T1_ID=CON

1. treatment 2 ID which will be used for naming files

T2_ID=STE OUT_DIR=/ebi/bscratch/jarrah/uqphilippb/2014_4_17_FixPAV/CSR_pipeline/

1. directory storing treatment 1 data files. The data files can be fasta or fastq formats.

DATA_DIR_T1=/ebi/bscratch/jarrah/uqphilippb/2014_4_17_FixPAV/CSR_pipeline/whole/whole_2.150.1sd/

1. directory storing treatment 2 data files. The data files can be fasta or fastq formats.

DATA_DIR_T2=/ebi/bscratch/jarrah/uqphilippb/2014_4_17_FixPAV/CSR_pipeline/truncated/truncated_2.150.1sd/

[ADVANCED]

1. Important setting in jellyfish for tuning jellyfish performance. A larger hash size, more memory will be used
2. but less sub-count files will be generated. Default is 10000000.

JELLYFISH_HASH_SIZE=10000000

1. Size of the jellyfish hash table. Use a size large enough to contain all of the K-mers such that 80% * s > number of distinct K-mers. [Default: 10000000].
2. jellyfish_hash_size = 16G
3. Length of counter in the hash table. See jellyfish manual for explanation. [Default: 4]

JELLYFISH_COUNTER_BITS = 18

1. Number of bases used for splitting the k-mer files into sub-files. A larger number reduces RAM usage. Maximum at the moment is 3. [Default: 2]

N_SPLIT_BASES = 3

1. Minimum occurrence count for a k-mer to be considered as a candidate Presence / Absence K-mer. [Default: 4]

MIN_OCC = 4<nowiki>

The command is:

   python pipeline.py 
        --s1 ./truncated/truncated_2.150.1sd/truncated_2.150.1sd.R1.fasta ./truncated/truncated_2.150.1sd/truncated_2.150.1sd.R2.fasta 
        --s2 ./whole/whole_2.150.1sd/whole_2.150.1sd.R1.fasta ./whole/whole_2.150.1sd/whole_2.150.1sd.R2.fasta 
        -c default.config  
        -o output_folder/

or, easier:

   python pipeline.py 
        --s1 ./truncated/truncated_2.150.1sd/truncated_2.150.1sd.R?.fasta
        --s2 ./whole/whole_2.150.1sd/whole_2.150.1sd.R?.fasta
        -c default.config  
        -o output_folder/

This will read the configuration from default.config and generate all output files in output_folder.

How to interpret the results?

• You can download the results of the sample data here.

Reference

• Marçais, G. and Kingsford, C. (2011) A fast, lock-free approach for efficient parallel counting of occurrences of k-mers, Bioinformatics, 27, 764-770.

Back to Main_Page