empty fastq files after Miseq is done

[mmhefny]

Dear all,
I am new in this field and I done my first sequencing using MiSeq. I checked the files and found that all generated fastq.gz files are empty except two files with names Undetermined_****.fastq.gz are in the Data/Intensities/BaseCalls/ folder which has data (each is about 200 kb).

Could you please help me to explain where is the problem?

Thank you

[GenoMax]

Sounds like you had a demultiplexing failure. Check your sample sheet one more time (easy mistake is to provide the indexes in rev-comp) and re-queue demultiplexing. You should use Illumina Experiment Manager software (windows only) available from Illumina to make the sample sheet, if you are new to sequencing.

At worst, your run may have failed. Best solution to check on either possibilities is to contact Illumina tech support. They should be able to take a look at your MiSeq remotely (if your MiSeq is not on network then they will ask for a few files) and diagnose.

[mmhefny]

Thank you.

I will check both possibilities. I looked at the CompletedJobInfo file and it seems to be fine. I attached also my samplesheet file. I hope you have time to check them.

I really appreciate your efforts.

[mmhefny]

Sounds like you had a demultiplexing failure. Check your sample sheet one more time (easy mistake is to provide the indexes in rev-comp) and re-queue demultiplexing. You should use Illumina Experiment Manager software (windows only) available from Illumina to make the sample sheet, if you are new to sequencing.

At worst, your run may have failed. Best solution to check on either possibilities is to contact Illumina tech support. They should be able to take a look at your MiSeq remotely (if your MiSeq is not on network then they will ask for a few files) and diagnose.

Thank you.

I will check both possibilities. I looked at the CompletedJobInfo file and it seems to be fine. I attached also my samplesheet file. I hope you have time to check them.

I really appreciate your efforts.

[GenoMax]

It won't be very useful to look at the Samplesheet. If MiSeq accepted it at run time then it must have been in the correct format. If your samples did not demultiplex then it is very likely that you have wrong index sequence information in your samplesheet.

Can you post a screenshot of the Sequence Analysis Viewer info for this run? Especially the alignment to phiX. That would be useful for diagnosis.

[mmhefny]

It won't be very useful to look at the Samplesheet. If MiSeq accepted it at run time then it must have been in the correct format. If your samples did not demultiplex then it is very likely that you have wrong index sequence information in your samplesheet.

Can you post a screenshot of the Sequence Analysis Viewer info for this run? Especially the alignment to phiX. That would be useful for diagnosis.

Unfortunately due to COVID 19 measures I have no access to the MiSeq machine now. I only have the run files. Also we did not include phiX sample control. Is there any file in the run folder can be representative so that I share it with you?

I really appreciate your help.

[GenoMax]

Are you able to use command line? If you are I can suggest some options to look at those "Undetermined" files to see what may be going on.

If you are able to open them (on PC/Mac) post a small number of example reads here.

[jdk787]

If you look at the DemultiplexSummaryF1L1.txt it will tell you what barcodes were used on the run if the demultiplexing worked.

It's located at
<run folder>\Data\Intensities\BaseCalls\Alignment

[mmhefny]

Are you able to use command line? If you are I can suggest some options to look at those "Undetermined" files to see what may be going on.

If you are able to open them (on PC/Mac) post a small number of example reads here.

Yes I can. Here are some reads
@M02006:24:000000000-J397M:1:1101:17423:1618 2:N:0:0
CATTGTTCCTTTTGCTTCTTTCCTTTCCCTCTTCTCTTTCTTTTCTCTTCTCTCCTTCTTTTTTCTTTTTTGCCTCTCTTGTTTTCTTACTCTCTTTCCCCCTCTTTTTTCTCTTTCTTTCTTCTCCCTCTCTCTCCCTCACCTTTTCTAT
+
1111>3333B@3111A13D1331333111A0011A133333111212D212A210111111D1/011111/A1111B01A111002D12111211212110///01B11/012@21221B2212B1100000111100000/0111121B2
@M02006:24:000000000-J397M:1:1101:17369:2159 2:N:0:0
TCGCCGGCCTAGTAGGCCCTTCCCTTAGTCTCCTTCTTCTCGCTGCATTCTGACACCCCGGCTCTCTACTTGGCGCTGCTCAACTTTCTAATGTGGCCGCGCGTCGTGTAGGGCACTAGTGTCGTCCTGCGTGTCGATCTCGGAGGTCGCG
+
11111>>1100B1AG11B00A000111B22211B111D22B///B/A1222221B1///A////0112D211F0@//E/1111021FD2222B2BF/B>///>//</?/011?0//01?1?1</<//.<1>.<>0CA<..0<..--./:--
@M02006:24:000000000-J397M:1:1101:18125:2191 2:N:0:0
CCCCTTTCCCTCTGTCTTCTCCCTCGGCGCCTTCTTTGCTCTTCTCGCCGGCTTCGTTGTCCGCGCGTCGTGTCGTGCCCGCGTGTCCCGCTGGTTGTCGCTCTCTGTTGTCGCCGTCTCCTTCCCCCCCCGCCTGCCCTCCCCTTCTGTT
+
11>11111111>11B333331AA1BA0000A00111A111A1112BA/A//A//0>000B11//>////>///0///</0//////011/////01/?1/////1111<101<..-<-./00<0///<------...../....9.-////
@M02006:24:000000000-J397M:1:1101:21922:3716 2:N:0:0
GTTTTTCCTCGTCTGCCCGGCCTCCGTTCTTTCTTTCTCTTCTCCCCGTGCTGCTTCCCGCTTCTTCTCCTTTGCACCCCTCTTTCTTCCTGGTCCTCTTGGCGTGCGCCTTCTTGCCTCTTCCTCCCTGTCTCCCGTTTGTCCGCACGTG
+
1>111111311>11A1111000000BA100AA22D12BA2212110///B//1A11110/////11221111211210///?01122121@101111111110///?><///111111111111110/01<12>10//00/021/////0.
@M02006:24:000000000-J397M:1:1101:17741:4534 2:N:0:0
CTTCTCTTTTCTTTTCGGCGCTTGCTCCGGTCTCCTTGGCCCCTCCTTCCGTCTCCTGCTTCGCGCTGCCTTCGGTCGCCCCGGGCCCTTCCTTGGCGCTGGGCCGCGCGTCGTGTCGGGCCCGCGTGTCCCCGCGTGTGTCGATCTCGGT

[mmhefny]

If you look at the DemultiplexSummaryF1L1.txt it will tell you what barcodes were used on the run if the demultiplexing worked.

It's located at
<run folder>\Data\Intensities\BaseCalls\Alignment

Yes I can see this file. None of the barcodes which I used has hit and only some hits are for different barcodes

[GenoMax]

@mmhefny,

Are you sure this run was set up to run a index cycle? This run header does not seem to make it look so.

@M02006:24:000000000-J397M:1:1101:21922:3716 2:N:0:0

Normally one should see something like this at the end of fastq header for index sequence

@M02006:24:000000000-J397M:1:1101:21922:3716 1:N:0:GAACGCAA+CTGGCACT

Can you find the RunInfo.xml file and post the section that shows this part?

<Reads>
      <Read NumCycles="150" Number="1" IsIndexedRead="N" />
      <Read NumCycles="8" Number="2" IsIndexedRead="Y" />
      <Read NumCycles="8" Number="3" IsIndexedRead="Y" />
      <Read NumCycles="150" Number="4" IsIndexedRead="N" />
    </Reads>

[mmhefny]

@mmhefny,

Are you sure this run was set up to run a index cycle? This run header does not seem to make it look so.

@M02006:24:000000000-J397M:1:1101:21922:3716 2:N:0:0

Normally one should see something like this at the end of fastq header for index sequence

@M02006:24:000000000-J397M:1:1101:21922:3716 1:N:0:GAACGCAA+CTGGCACT

Can you find the RunInfo.xml file and post the section that shows this part?

<Reads>
      <Read NumCycles="150" Number="1" IsIndexedRead="N" />
      <Read NumCycles="8" Number="2" IsIndexedRead="Y" />
      <Read NumCycles="8" Number="3" IsIndexedRead="Y" />
      <Read NumCycles="150" Number="4" IsIndexedRead="N" />
    </Reads>

This part of RunInfo.xml is as follow

    <Reads>
      <Read NumCycles="151" Number="1" IsIndexedRead="N" />
      <Read NumCycles="8" Number="2" IsIndexedRead="Y" />
      <Read NumCycles="8" Number="3" IsIndexedRead="Y" />
      <Read NumCycles="151" Number="4" IsIndexedRead="N" />
    </Reads>

[GenoMax]

Ah so your run was dual-indexed. I am puzzled as to why your reads don't have any index sequence in fastq headers.

What do you see in DemuxSummaryF1L1.txt?

Scroll down the file until your see this section:

### Most Popular Unknown Index Sequences
### Columns: Index_Sequence Hit_Count
CTCATCAC+GGTGGCAC       2940
AGGTCAAG+CGCGTAAT       860
TCCAGGTA+GCTTCAGT       700

[mmhefny]

Ah so your run was dual-indexed. I am puzzled as to why your reads don't have any index sequence in fastq headers.

What do you see in DemuxSummaryF1L1.txt?

Scroll down the file until your see this section:

### Most Popular Unknown Index Sequences
### Columns: Index_Sequence Hit_Count
CTCATCAC+GGTGGCAC       2940
AGGTCAAG+CGCGTAAT       860
TCCAGGTA+GCTTCAGT       700

they are:

### Most Popular Index Sequences
### Columns: Sequence ReverseComplement HitCount
Index
C.CCC.CC GG.GGG.G 96
C.CCT.GC GC.AGG.G 94
C.CGC.CG CG.GCG.G 78
C.GCC.TT AA.GGC.G 65
C.ACC.CT AG.GGT.G 44
C.CCC.CT AG.GGG.G 40
C.ACG.GG CC.CGT.G 38
C.CGC.GG CC.GCG.G 37
C.AGC.GG CC.GCT.G 36
C.CGC.TC GA.GCG.G 33
C.CCC.CG CG.GGG.G 29

[mmhefny]

Ah so your run was dual-indexed. I am puzzled as to why your reads don't have any index sequence in fastq headers.

What do you see in DemuxSummaryF1L1.txt?

Scroll down the file until your see this section:

### Most Popular Unknown Index Sequences
### Columns: Index_Sequence Hit_Count
CTCATCAC+GGTGGCAC       2940
AGGTCAAG+CGCGTAAT       860
TCCAGGTA+GCTTCAGT       700

This is the file and it does not have this Unknown Index Sequences