FaQCs: Quality Control of Next Generation Sequencing Data

PREREQUISITES

1. The main program is developed in Perl v 5.8.8.
2. Parallel::ForkManager module from CPAN
   (http://search.cpan.org/~dlux/Parallel-ForkManager-0.7.9/lib/Parallel/ForkManager.pm)
3. String::Approx module from CPAN
   (http://search.cpan.org/~jhi/String-Approx-3.27/Approx.pm)
4. R for ploting
   (http://www.r-project.org/)
5. Jellyfish for kmer counting (Optional) (http://www.cbcb.umd.edu/software/jellyfish/)

Note: The two Perl modules can be installed by INSTALL.sh script in the lib directory.

cd lib
./INSTALL.sh

---

BASIC USAGE

• Trimming by quality 5 and filtering reads with any ambiguous base or low complexity.

  $ perl FaQCs.pl -p reads1.fastq reads2.fastq -d out_directory

• Quailty check only on subsamples of input, no trimming and filtering.

  $ perl FaQCs.pl -p reads1.fastq reads2.fastq -d out_directory -qc_only

---

Full USAGE

Usage: perl FaQCs.pl [options] [-u unpaired.fastq] -p reads1.fastq reads2.fastq -d out_directory
Version 1.34
Input File: (can use more than once)
        -u            <Files> Unpaired reads
        
        -p            <Files> Paired reads in two files and separate by space
Trim:
        -mode         "HARD" or "BWA" or "BWA_plus" (default BWA_plus)
                      BWA trim is NOT A HARD cutoff! (see bwa's bwa_trim_read() function in bwaseqio.c)

        -q            <INT> Targets # as quality level (default 5) for trimming

        -5end         <INT> Cut # bp from 5 end before quality trimming/filtering 
  
        -3end         <INT> Cut # bp from 3 end before quality trimming/filtering 

        -adapter      <bool> Trim reads with illumina adapter/primers (default: no)
                      -rate   <FLOAT> Mismatch ratio of adapters' length (default: 0.2, allow 20% mismatches)
        					
        -artifactFile  <File>    additional artifact (adapters/primers/contaminations) reference file in fasta format 
Filters:
        -min_L        <INT> Trimmed read should have to be at least this minimum length (default:50)

        -avg_q        <NUM> Average quality cutoff (default:0, no filtering)
        
        -n            <INT> Trimmed read has more than this number of continuous base "N" will be discarded. 
                      (default: 2, "NN") 

        -lc           <FLOAT> Low complexity filter ratio, Maximum fraction of mono-/di-nucleotide sequence  (default: 0.85)

        -phiX         <bool> Filter phiX reads (slow)
        
Q_Format:
        -ascii        Encoding type: 33 or 64 or autoCheck (default)
                      Type of ASCII encoding: 33 (standard) or 64 (illumina 1.3+)

        -out_ascii    Output encoding. (default: 33)
Output:
        -prefix       <TEXT> Output file prefix. (default: QC)

        -stats        <File> Statistical numbers output file (default: prefix.stats.txt)

        -d            <PATH> Output directory.
Options:
        -t            <INT > # of CPUs to run the script (default:2 )

        -split_size   <INT> Split the input file into several sub files by sequence number (default: 1000000) 

        -qc_only      <bool> no Filters, no Trimming, report numbers.

        -kmer_rarefaction     <bool>   
                      Turn on the kmer calculation. Turn on will slow down ~10 times. (default:Calculation is off.)
                      (meaningless if -subset is too small)
                      -m  <INT>     kmer for rarefaction curve (range:[2,31], default 31)

        -subset       <INT>   Use this nubmer x split_size for qc_only and kmer_rarefaction  
                              (default: 10,  10x1000000 SE reads, 20x1000000 PE reads)

        -discard      <bool> Output discarded reads to prefix.discard.fastq (default: 0, not output)

        -substitute   <bool> Replace "N" in the trimmed reads with random base A,T,C ,or G (default: 0, off)

        -trim_only    <bool> No quality report. Output trimmed reads only.

        -5trim_off    <bool> Turn off trimming from 5'end.

        -debug        <bool> keep intermediate files

---

VERSION HISTORY

======== Version 1.34

• add option "-5trim_off Turn off trimming from 5'end."
• add INSTALL.sh script for two requried perl modules installations.

======== Version 1.33

• input paired no need quote for exploit the autocomplete feature
• add trim_only option
• mode with "HARD" or "BWA" or "BWA_plus" (default BWA_plus)

======== Version 1.32

• add -5end and -3end to cut x number base from 5' end or 3' end before quality trimming/filtering
• fix bug on phiX filtering with reverse complementary strand hit
• fix error when all reads in subsample are filtered/trimmed.

======== Version 1.31

• report raw subsample graphic results side-by-side with qc results for comparison.

======== Version 1.3

• add -phiX to filter phiX reads
• add -substitute to replace "N" in the trimmed reads with random base A,T,C ,or G
• change -adapter behavior from filtering to trimming
• change -n behavior from # of tolerance to number of continuous base "N" filtering

======== Version 1.2

• add -adapter and -artifactFile for filtering reads with Adapters/Primers and other contaminations
• require String::Approx module from CPAN for above function

======== Version 1.1 New features and changes in illumina_fastq_qc version 1.1 with respect to version 1.0:

• add -qc_only option for quick quality check without trimming and filtering
• add -discard option to output discarded reads

======== Version 1.0 Stable function release. Features:

• trim bidirection
• minimium length filtering after trim
• "N" base filter
• low complexity filter
• average read quality filter
• autocheck quality encoding and quality encoding coversion
• multi-threads (required Parallel::ForkManager)
• input paired end reads aware

---

CITATION

Chienchi Lo, PatrickS.G. Chain (2014) Rapid evaluation and Quality Control of Next Generation Sequencing Data with FaQCs. BMC Bioinformatics. 2014 Nov 19;15