[ymc]

http://www.illumina.com/systems/nextseq-sequencer/kits.html

I find a NextSeq v2 kit here. Is it something new?

[kentawan]

http://www.illumina.com/systems/next...ncer/kits.html

I find a NextSeq v2 kit here. Is it something new?

I just gave my local distributor a call. He said that this kit will be ready for shipment on February 2015. Pricing will be the same as the v1 kits!

Finally some hope for NextSeq 500 users!

[Elsie]

Hi Brian,

Thanks so much for this. I am trying to repeat your above commands, using interleaved files, and I get this error, can you help?
Thanks.

bbmap.sh maxindel=200 in=trimmed.fq.gz mhist=mhist.txt bhist=bhist.txt qhist=qhist.txt qahist=qahist.txt
java -Djava.library.path=/bbmap/jni/ -ea -Xmx43110m -cp /bbmap/current/ align2.BBMap build=1 overwrite=true fastareadlen=500 maxindel=200 in=trimmed.fq.gz mhist=mhist.txt bhist=bhist.txt qhist=qhist.txt qahist=qahist.txt
Executing align2.BBMap [build=1, overwrite=true, fastareadlen=500, maxindel=200, in=trimmed.fq.gz, mhist=mhist.txt, bhist=bhist.txt, qhist=qhist.txt, qahist=qahist.txt]

BBMap version 34.56
Set match histogram output to mhist.txt
Set base content histogram output to bhist.txt
Set quality histogram output to qhist.txt
Set quality accuracy histogram output to qahist.txt
Retaining first best site only for ambiguous mappings.
No output file.
Exception in thread "main" java.lang.RuntimeException: Can't find file ref/genome/1/summary.txt
at fileIO.ReadWrite.getRawInputStream(ReadWrite.java:815)
at fileIO.ReadWrite.getInputStream(ReadWrite.java:780)
at fileIO.TextFile.open(TextFile.java:277)
at fileIO.TextFile.<init>(TextFile.java:94)
at dna.Data.setGenome2(Data.java:839)
at dna.Data.setGenome(Data.java:785)
at align2.BBMap.loadIndex(BBMap.java:302)
at align2.BBMap.main(BBMap.java:32)

[Brian Bushnell]

Hi Elsie,

You have to index the reference first. For example:

bbmap.sh ref=genome.fasta

Wait for that to finish, then map.

-Brian

[Elsie]

Thanks Brian, unfortunately I do not have a reference for this sequence!, so I'm assuming no way around this?

[Brian Bushnell]

The only way around is to try to assemble it first. If it's a bacteria, and you have sufficient coverage, you can get a decent assembly in a few minutes with Velvet. BBMap will not work without an assembly, but it doesn't have to be a good assembly - a quick one with short contigs is fine for this purpose, as long as those contigs are several times larger than read length.

[Elsie]

Thank you Brian, that is really helpful, and incredibly prompt. Thank you so much.

[Brian Bushnell]

You're welcome - let us know if you discover anything interesting!

[Elsie]

Hi Brian,
sorry, still having issues. I've now switched to some NextSeq data generated with mouse and human data. I'm getting Nas in my histograms, I think there is something wrong with my index, info.txt gives me:
#Chromosome sizes
#Generated on Fri Mar 06 21:58:51 EST 2015
#Version 5
#chrom scaffolds contigs length defined undefined startPad stopPad
1 4 41 493337098 479857220 13479878 8000 8000
2 5 61 512852808 496275279 16577529 8000 8000
3 3 70 439025362 428441699 10583663 8000 8000
4 3 78 468391081 456374140 12016941 8000 8000
5 3 46 424587819 413803382 10784437 8000 8000
6 4 155 387396307 372786010 14610297 8000 8000
What happened to the other chromosomes? I must be doing something wrong but I am just doing the bbmap ref command as indicated previously.
thanks.

[GenoMax]

Hi Brian,
sorry, still having issues. I've now switched to some NextSeq data generated with mouse and human data. I'm getting Nas in my histograms, I think there is something wrong with my index

What is Nas? (N's?)

The index should be ok. I think Brian is concatenating all chromosomes and then creating the index so that file is not a literal equivalent of human/mouse genome (file I have looks similar to yours).

Are you getting an error when you do the mapping?

[fanli]

Hi all,

Just wanted to add our data as well - this was from an RNA-seq library and I don't have paired HiSeq data. Still, you can see some ambitious reporting of quality scores albeit not nearly as bad as what aeonsim showed.

We're hopefully going to run a v2 kit soon and I'll update with those stats when I get them!

[Brian Bushnell]

The index should be ok. I think Brian is concatenating all chromosomes and then creating the index so that file is not a literal equivalent of human/mouse genome (file I have looks similar to yours).

That's correct. They're called chromosomes for legacy reasons (the chunks used to be one real chromosome each) but it's more efficient to pack them.

[aeonsim]

Hi all,

Just wanted to add our data as well - this was from an RNA-seq library and I don't have paired HiSeq data. Still, you can see some ambitious reporting of quality scores albeit not nearly as bad as what aeonsim showed.

We're hopefully going to run a v2 kit soon and I'll update with those stats when I get them!

I'd acctually say they're worse than what we had, considering your using PE80bp and the first 10 or so bases on the forward reads shows an average Quality score drop of ~10 on the Phred Scale (~30 to 20).

However our conculsion from our testing is that the NextSeq with V1 chemistry is ok for RNAseq as the reads still map fine and the coverage is high, it's however not suitable for variant calling especially when one is interested in de novo variants or low coverage. As a result it's only being used internally for RNAseq currently.

We will aparently get access to the V2 kits as soon as they're available to see if that fixes the issue.

[Elsie]

No error, just Nas which is why I think I am missing something:
reformat.sh in1=R1.fastq in2=R2.fastq out=interleaved
gzip interleaved
bbmap.sh ref=hg19.fa
bbduk.sh in=interleaved.gz out=trimmed.fq.gz ktrim=r k=23 hdist=1 mink=11 tpe tbo minlen=90 ref=truseq.fa.gz,nextera.fa.gz
bbmap.sh maxindex=200 in=trimmed.fq.fq mhist=mhist.txt bhist=bhist.txt qhist=qhist.txt qahist=qahist.txt

BBMap version 34.56
Set match histogram output to mhist.txt
Set base content histogram output to bhist.txt
Set quality histogram output to qhist.txt
Set quality accuracy histogram output to qahist.txt
Retaining first best site only for ambiguous mappings.
No output file.
Set genome to 1

Loaded Reference: 5.025 seconds.
Loading index for chunk 1-7, build 1
Generated Index: 6.192 seconds.
Analyzed Index: 7.512 seconds.
Cleared Memory: 0.461 seconds.
Processing reads in single-ended mode.
Started read stream.
Started 16 mapping threads.
Detecting finished threads: 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15

------------------ Results ------------------

Genome: 1
Key Length: 13
Max Indel: 200
Minimum Score Ratio: 0.56
Mapping Mode: normal
Reads Used: 0 (0 bases)

Mapping: 0.447 seconds.
Reads/sec: 0.00
kBases/sec: 0.00


Read 1 data: pct reads num reads pct bases num bases

mapped: NaN% 0 NaN% 0
unambiguous: NaN% 0 NaN% 0
ambiguous: NaN% 0 NaN% 0
low-Q discards: NaN% 0 NaN% 0

perfect best site: NaN% 0 NaN% 0
semiperfect site: NaN% 0 NaN% 0

Match Rate: NA NA NaN% 0
Error Rate: NaN% 0 NaN% 0
Sub Rate: NaN% 0 NaN% 0
Del Rate: NaN% 0 NaN% 0
Ins Rate: NaN% 0 NaN% 0
N Rate: NaN% 0 NaN% 0

Total time: 19.975 seconds.

Any advice greatly appreciated. thanks.

[GenoMax]

@Elsie: Are you trying to analyze NextSeq500 data or just creating stats? This thread was originally about quality of NextSeq500 reads and the procedure that Brian had posted was to create stats files (not actual alignments).

If you are actually trying to analyze real data then there is no need to create interleaved data sets. You can directly trim and then align R1/R2 reads against human genome. You need to specify an output file to store the aligned reads.

A minimal command line for doing the mapping would be following. More examples in the BBMap thread: http://seqanswers.com/forums/showthread.php?t=41057

$ bbmap.sh -Xmx30g in=trimmedfq.gz path=/path_to_BBMap_index_top_folder_with_ref_directory/ out=sample_ID.sam qin=33

Change the path to BBMap index according to your local path. Add additional options (there are plenty) as needed depending on kind of experiment you are analyzing.