najoshi / sabre Goto Github PK

Hi,
new to this biof stuff. But used sabre a few times on my fastqs and worked great (Thanks a Lot). However, last run....can use bowtie2 to map non demultiplexed fastq fine no errors. Demultiplex with sabre then about half the new fastqs throw errors saying "Error: Read M00733:9:000000000-A5T8E:1:2116:15800:22597 2:N:0:1 has more read characters than quality values.
terminate called after throwing an instance of 'int'
bowtie2-align died with signal 6 (ABRT) (core dumped)"

Got MiSeq to remake fastqs and tried again. Same problem. Deleted the entry for the error throwing line, found a new one. Gave up after deleting four reads worth.

Not sure what to do next!!!

Any help appreciated..

~/sabre-master$ make
gcc -Wall -pedantic -DVERSION=1.00 -O3 -c src/barcode.c
FATAL:/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/../libexec/as/x86_64/as: I don't understand 'm' flag!
make: *** [barcode.o] Error 1

Any suggestions?
I also heard that sable requires something called zlib - is this the reason for this error? Where and how do I get zlib?

Thanks!

This is needed for packaging.

Hi, how should one incorporate technical replicates of the same individual sample within a dataset into their saber analysis? Specifically, how is this coded? To provide an example, in ipyrad's demultiplexing script (here: https://github.com/dereneaton/ipyrad/blob/master/ipyrad/assemble/demultiplex.py), technical replicates must be named <sample_name>"-technical-replicate-". I'm wondering if something like this is possible with sabre.

Thanks for your help. Let me know if you need further information.

~J

Hi, my index it at the end of my reads..
Is it working ? It seems not

Hi,
Is there a way to run sabre in parallel, and control the number of cores / threads used ?
Thanks

Hello!

I am having a problem lately. I have a script that used to work before (maybe I changed something that I cannot remember...) but since today I get an error message. The script just does some steps (that work fine) and then calls sabre like this:

sabre pe -f ${demux_dir}/temp_S0_R1_001.fastq -r ${demux_dir}/temp_S0_R2_001.fastq -b $sample_names_file -u $unmatched_R1 -w $unmatched_R2 -c -m 1

However, it crashes every time throwing this error (as seen in the log files):

demuxbysabre.sh: line 38: 15825 Segmentation fault (core dumped) sabre pe -f ${demux_dir}/temp_S0_R1_001.fastq -r ${demux_dir}/temp_S0_R2_001.fastq -b $sample_names_file -u $unmatched_R1 -w $unmatched_R2 -c -m 1

Not sure what could be the problem!! I installed sabre through conda, btw. Not sure if that matters. Thank you!

Hi,
I'm Hanwen. I saw the usage about sabre on website (https://astrobiomike.github.io/amplicon/demultiplexing). And it is mentioned in the end of this document that the data with one index can be demultiplexed. This is the key problem in my current analysis project and I don't know how to address it. Could you help me to do it?

FYI

https://github.com/tseemann/homebrew-bioinformatics-linux/blob/master/sabre.rb

When trying to install on Ubuntu 13.04 with default zlib installed I get the error:

jstantongeddes@AntLab:/opt/software/sabre$ sudo make
gcc -Wall -pedantic -DVERSION=1.00 -lz -O3 barcode.o demulti_single.o demulti_paired.o sabre.o -o sabre
demulti_paired.o: In function `ks_getuntil':
demulti_paired.c:(.text+0x78): undefined reference to `gzread'
demulti_paired.o: In function `kseq_read':
demulti_paired.c:(.text+0x48d): undefined reference to `gzread'
demulti_paired.c:(.text+0x4dc): undefined reference to `gzread'
demulti_paired.c:(.text+0x60a): undefined reference to `gzread'
demulti_paired.c:(.text+0x687): undefined reference to `gzread'
demulti_paired.o: In function `paired_main':
demulti_paired.c:(.text+0xb67): undefined reference to `gzopen'
demulti_paired.c:(.text+0xb82): undefined reference to `gzopen'
demulti_paired.c:(.text+0x1249): undefined reference to `gzclose'
demulti_paired.c:(.text+0x1253): undefined reference to `gzclose'
demulti_single.o: In function `ks_getuntil':
demulti_single.c:(.text+0x78): undefined reference to `gzread'
demulti_single.o: In function `single_main':
demulti_single.c:(.text+0x57e): undefined reference to `gzopen'
demulti_single.c:(.text+0x868): undefined reference to `gzread'
demulti_single.c:(.text+0x8b8): undefined reference to `gzread'
demulti_single.c:(.text+0xc17): undefined reference to `gzread'
demulti_single.c:(.text+0xca8): undefined reference to `gzread'
demulti_single.c:(.text+0xd78): undefined reference to `gzclose'
collect2: ld returned 1 exit status
make: *** [build] Error 1

I did find a solution on stackoverflow that allowed me to successfully install...but I'm confused why this wouldn't just work by default.

Thanks for making the program available!

Hi,

The output is a bit weird, This the kind of reads I have at the end:

@WINDU:150:CA3H4ANXX:4:1101:1208:2116 1:N:0:CTTGTA
ATATCCTTGGGCATGATGGTGACGCGCTTGGCGTGGATGGCGCACAAGTTGGTGTCCTCGAAGAGGCCGACCAGGTAAGCCTCGCTGGCCTCCT
+WINDU:150:CA3H4ANXX:4:1101:1208:2116 1:N:0:CTTGTA
FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF/FFBFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF

While the 3rd line should have only a '+' , why the name is copied ?

Hi,

I used the Sabre to demultiplex data but no matter if I add -c parameter or not, the output demultiplexed fastq files are only 0 reads.

The first column is the size of the file which are all 0.

But the unmached R1 amd R2 fastq files seem large:

Does this mean there is something wrong with my barcode mapping file, or can be some parameter issues that I have not addressed yet?

Thanks!
Leran

Is there a hard limit to the number of samples Sabre PE can demultiplex? We have 142 samples in a barcode file- this causes Sabre PE to segfault 11 and fail. If we drop below 124 samples, the program runs successfully. Is this something hard-coded, or an option that can be changed prior to compilation to allow for more samples within a single demultiplexing run?

Hi, I am wondering what method this software uses to match the read barcodes to that provided? I am running it with different levels of mismatching (0,1, 2), but when I go from 1 to 2 the allocation of reads to certain barcodes is decreasing. My barcodes have been designed to be identifiable with up to 3 mismatches, so I was curious what could cause this and was wondering how the software selects the "correct" barcode if you allow mismatches?

Hi Nik,

We're moving to the use of combinatorial barcoding to allow multiplexing more than 96 samples per lane. This involves different R1 & R2 barcodes (e.g. 12 R1 barcodes * 8 R2 barcodes = 96 unique barcode combinations).

While sabre can trim barcodes from both pairs, from what I can tell from the code it does no matching on the R2 read.

Therefore I propose (and am happy to implement myself, assuming it will be merged) the following:

• A new command, sabre comb, to handle this use case.
• A new barcode file format to handle this:

....

Cheers,
Kevin

Hi!
I used sabre for demultiplexing my reads and would like to cite your tool. I can not find information about how to cite it. Should I just refer to the git page here or you have a paper to cite?
Thanks

Hi,
I am trying to add sabre to ~/.profile but it does not work.
kindly help