Trimmomatic not able to remove low sequence bases

Question

[David_emir]

Hello Friends,

I am dealing with an RNA-Seq Paired end data. My goal is to identify Differential expressed genes. While preprocessing the data I wanted to remove the . I used Trimmomatic command to filters out paired-end reads (PE) whose mean base quality is below 20 (AVGQUAL:20).

First I executed command which removes the Adapter sequence, It did remove the seq.

java -jar trimmomatic-0.33.jar PE -phred64 SRR1600265_1.fastq SRR1600265_2.fastq R1_pairedout.fastq R1_unpairedout.fastq R2_pairedout.fastq R2_unpairedout.fastq ILLUMINACLIP:/home/NSCLC/Trimmomatic-0.33/adapters/TruSeq3-PE-2.fa:2:30:10:1:trueTrimmomaticPE: Started with arguments: -phred64 SRR1600265_1.fastq SRR1600265_2.fastq R1_pairedout.fastq R1_unpairedout.fastq R2_pairedout.fastq R2_unpairedout.fastq ILLUMINACLIP:/home/NSCLC/Trimmomatic-0.33/adapters/TruSeq3-PE-2.fa:2:30:10:1:true

Multiple cores found: Using 16 threads
Using PrefixPair: 'TACACTCTTTCCCTACACGACGCTCTTCCGATCT' and 'GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT'
Using Long Clipping Sequence: 'AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGTA'
Using Long Clipping Sequence: 'AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC'
Using Long Clipping Sequence: 'GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT'
Using Long Clipping Sequence: 'TACACTCTTTCCCTACACGACGCTCTTCCGATCT'
ILLUMINACLIP: Using 1 prefix pairs, 4 forward/reverse sequences, 0 forward only sequences, 0 reverse only sequences

Input Read Pairs: 43254901 Both Surviving: 42693867 (98.70%) Forward Only Surviving: 1817 (0.00%) Reverse Only Surviving: 467152 (1.08%) Dropped: 92065 (0.21%)
TrimmomaticPE: Completed successfully

NOW the real challenge:

I am trying with Trimmomatic command for paired-end reads (PE) trims bases from the 3' end when the base quality is below 20 (TRAILING:20) and filters out reads which are shorter than 50 bases after trimming(MINLEN:50).

# java -jar trimmomatic-0.33.jar PE -phred64 R1_pairedout.fastq R2_pairedout.fastq R1_split_pairedout.fastq R1_split_unpairedout.fastq R2_split_pairedout.fastq R2_split_unpairedout.fastq LEADING:5 TRAILING:5 AVGQUAL:20

TrimmomaticPE: Started with aarguments: -phred64 R1_pairedout.fastq R2_pairedout.fastq R1_split_pairedout.fastq R1_split_unpairedout.fastq R2_split_pairedout.fastq R2_split_unpairedout.fastq LEADING:5 TRAILING:5 AVGQUAL:20

Multiple cores found: Using 16 threads
Input Read Pairs: 42693867 Both Surviving: 0 (0.00%) Forward Only Surviving: 0 (0.00%) Reverse Only Surviving: 0 (0.00%) Dropped: 42693867 (100.00%)

I am not able to understand why trimmomatic not able to trim my reads. Please let me know your opinion.I might sound stupid, but I am Learning things. Thanks for your kind help

P.S: I started with SRA file paired end file --> Split it into reverse and forward bases --> run FASTQC on individual reads --> check for sequence quality for individual reads--> trim/filter for low bases using trimmomatic (for individual file each time.)..Let me know if this flow is correct.

[Gabriel R.]

"While preprocessing the data i wanted to remove the ."

the adapters ?

[lelle]

Are you sure phre64 is correct? Illumina is not using this anymore and I am not sure if the SRA tools ever used it.

Answer 1

[rtliu]

I have checked SRR1600265 fastq file with fastqc, it was Sanger /Illumina 1.9 encoding, you should use parameter -phred33

[Bioinformatics_NewComer]

How about let the tool [auto] detect phred score? Not sure if that's a correct way.

[rtliu]

Auto detecting phred score is usually working, but it is not 100% guaranteed.

[Bioinformatics_NewComer]

Ah. Thanks for sharing this info.
Could you share any instance where auto phred detection in Trimmomatic would fail?