Quality control on 454 data

Question

[deepkumar1983]

Dear Friends,

I am new to NGS field and after studing alot of literature on qualty control. I use trimmomatic, fastx and qtrim software for 454 RNA-seq data. I am attaching the fastqc output for that. But I am not happy with the per base sequence quality graph, per base GC content and per base sequence content. Any help will be appreciated.

Thanks
Deepak

[GouthamAtla]

What is your quality control pipeline?

[deepkumar1983]

I used trimmomatic all default option with headcrop 15. Then used that output as input for fastx tool kit with fastx_trmimmer and trim 8 base from the end of read. Next with fastq quality trimmer option is q 20 and p 30 remove low quality base. Finaaly used qtrim.

[User 59]

You didn't actually attach anything for anyone to look at...

[deepkumar1983]

Hi Swan this is my output FastQC report. Please help me how to correct these errors.

[GouthamAtla]

Where is the report ? Why don't u just give a Dropbox link ?

[deepkumar1983]

Hi this is the link of the file: https://www.dropbox.com/s/h6q97psk3gadf3h/Outputfile_fastqc.html?dl=0

My Pipeline Commands are

java \
  -jar /opt/software/Trimmomatic-0.32/trimmomatic-0.32.jar SE \
  SRR1646514.fastq \
  SRR1646514_filter1.fastq \
  ILLUMINACLIP:/opt/software/Trimmomatic-0.32/adapters/TruSeq2-SE.fa:2:30:10 \
  LEADING:3 \
  TRAILING:3 \
  SLIDINGWINDOW:4:15 \
  MINLEN:36 \
  HEADCROP:15

~/software/fastx_toolkit_0.0.13/fastx_trimmer \
  -Q 33 \
  -m 50 \
  -t 9 \
  -i SRR1646514_trimmomatics_filter.fastq \
  -o SRR1646514_trimmomatics_filter_t_9_m_50.fastq

QTrim_v1_1/QTrim_v1_1 \
  -m 30 \
  -mode 2 \
  -l 50 \
  -out_format 2 \
  -seq_id_stat \
  -plot pdf \
  -fastq ~/Desktop/SRR1646514_trimmomatics_filter.fastq

Answer 1

[Brian Bushnell]

Most trimming software is designed for Illumina data. I suggest you download BBDuk, and use this command:

bbduk.sh in=reads.fq out=trimmed.fq ref=adapters.fa k=23 ktrim=r mink=11 edist=1 qtrim=rl trimq=15

That should provide substantially better output, as it uses optimal dual-ended quality-trimming and allows indels in the adapter sequence, which is important in 454 data.

Answer 2

[Maxime B]

You can try Prinseq if Trimmomatic doesn't suit you http://edwards.sdsu.edu/cgi-bin/prinseq/prinseq.cgi

For your data it would be something like:

perl prinseq-lite.pl -verbose -fastq SRR1646514.fastq -stats_all -graph_data test.gd -custom_params "AGATCGGAAGAGCTCGTATGCCGTCTTCTGCTTG 1;AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGT 1; AGATCGGAAGAGCGGTTCAGCAGGAATGCCGAG 1" -min_len 36 -YourOtherTrimmingOptions

perl prinseq-graphs.pl -i test.gd -html_all -o test