fastq_quality_trimmer error

[shuang]

I'm having problems in running commands of fastx tool. For example, when I ran fastx_trimmer,
fastx_trimmer -i in.txt -o out.fastq

I received this error
fastx_trimmer: Invalid quality score value (char '#' ord 35 quality value -29) on line 4

Even I tried to add "-Q 64", still the same error.
fastx_trimmer -i in.txt -Q 64 -o out.fastq


Why is that??

[GenoMax]

If your data is in sanger format then -Q 33 would be the option to add.

[shuang]

Sorry, I forgot to mention my sample is Illumina. The sample length is 17kb.

[kmcarr]

I'm having problems in running commands of fastx tool. For example, when I ran fastx_trimmer,
fastx_trimmer -i in.txt -o out.fastq

I received this error
fastx_trimmer: Invalid quality score value (char '#' ord 35 quality value -29) on line 4

Even I tried to add "-Q 64", still the same error.
fastx_trimmer -i in.txt -Q 64 -o out.fastq


Why is that??

Illumina stopped using the Phred+64 offset sometime ago; they now use Phred+33. The fastx toolkit assumes Phred+64 by default so your two examples were exactly the same. As GenoMax said you need to use -Q33.

[shuang]

When I use -Q 33, it didn't go much further, neither. Below is the command and the error message.
fastx_trimmer -i in.txt -Q 33 -o all.fastq


fastx_trimmer: Error: invalid quality score data on line 448 (quality_tok = "??????B?D?DDDBB?FFFF;FFBB>FFHHDD?EGH>EFEAECEDHHHHHGHFE:FDGA-CEDHHDBF>AAACF=AAEBGFGF?CDDEHH+=CCCCBDAFF?DF.6BFF66...=B<*645,,5A,5,,53>?;>A,A5==*..8A?*:C"

[GenoMax]

shuang: Can you tell us what OS you are using to run the fastx_trimmer?

Did you open/edit the sequence file ("in.txt") in a non-unix OS (e.g. windows/OS X)?

What do you mean by sample length is 17kb?

[shuang]

I use Ubuntu. The in.txt is a fastq file, which was downloaded directly from the sequencing facility website.

17kb is the expected fully assembled length.

[dGho]

same problem

I am using fastq_quality_ trimmer for the first time and I am having this same problem also on Illumina generated data. Does anyone know what the solution to this is?

[JackieBadger]

Just use TRIMMOMATIC.
Works better with more options

[TonyBrooks]

There should be no space between the Q and the 33.

You also need to specify how you want to trim by setting variables for -f and -l e.g.

fastx_trimmer -Q33 -f 1 -l 50 -i input.fastq -o output.fastq

AFAIK Fastq's also need to be uncompressed for input but can be compressed on output using the -z flag.

Use fastx_trimmer -h for help

[yifangt]

I had the exactly same problem! It seems the problem is more of -Q33 option as I tried it.
Two things here:

1) which is the wrong quality score, i.e. ASCII char when you have the -Q33 on?

2) our sequence machine was run with the MS-Windows platform and the data was deposited on Linux storage disk. There might be a issue with this cross platform situation, but how to solve it?

Thanks a lot!