Mapping pair-end reads with nvBowtie

[psteinb]

Hi,
I'd like to map pair-ended reads to hg19 with nvBowtie. My bowtie2 call looks like this:

bowtie2 --rg-id SRR891269 --rg SRR891269 -p 10 -x /path/to/hg19 -X 2000 -1 /path/to/SRR891269_GM12878_50000cells_1.fastq -2 /path/to/SRR891269_GM12878_50000cells_2.fastq -S SRR891269.sam 

looking through the help section of nvBowtie, I am not quite clear how to call it accordingly. I already created the index of hg19 with nvBWT (as you might guess from prior issues).

Your help would be appreciated.
Best -

[jpantaleoni]

At the moment the command line differs from that of bowtie2, we'll likely
fix this in a next release.

For now a call looks like this:

nvBowtie --pe --fr --file-ref hg19 1.fq 2.fq output.sam

Basically with --pe the last 4 options are { index, mate1, mate2, output }.
On Apr 30, 2014 4:16 AM, "Peter Steinbach" notifications@github.com wrote:

Hi,
I'd like to map pair-ended reads to hg19 with nvBowtie. My bowtie2 call
looks like this:

bowtie2 --rg-id SRR891269 --rg SRR891269 -p 10 -x /path/to/hg19 -X 2000 -1 /path/to/SRR891269_GM12878_50000cells_1.fastq -2 /path/to/SRR891269_GM12878_50000cells_2.fastq -S SRR891269.sam

looking through the help section of nvBowtie, I am not quite clear how to
call it accordingly. I already created the index of hg19 with nvBWT (as you
might guess from prior issues).

Your help would be appreciated.
Best -

—
Reply to this email directly or view it on GitHubhttps://github.com/NVlabs/nvbio/issues/6
.

[psteinb]

Hello again,

I ran nvbowtie of these reads:
ftp://ftp-trace.ncbi.nlm.nih.gov/sra/sra-instant/reads/ByExp/sra/SRX%2FSRX298%2FSRX298000/SRR891268/

against the hg-19 genome through this command:

nvBowtie --pe --fr --file-ref hg19-index SRR891268_GM12878_50000cells_1.fastq SRR891268_GM12878_50000cells_2.fastq RR891268.sam

nvBowtie crashes with an error message that is unclear to me:

info    : aligning reads [170393600, 170917887]
verbose :   524288 reads
verbose :   52.429 M bps (75.0 MB)
verbose :   50.0 bps/read (min: 50, max: 50)
error   : opposite alignment kernel: unspecified launch failure
terminate called after throwing an instance of 'thrust::system::system_error'
  what():  unspecified launch failure

Best -

[jpantaleoni]

Hi Peter,

could you please tell us exactly how to obtain the files:

SRR891268_GM12878_50000cells_1.fastq
SRR891268_GM12878_50000cells_2.fastq

I have been downloading the .sra archive from the above directory and used
fastq-dump on it, which gave me:

SRR891268_1.fastq
SRR891268_2.fastq

is this equivalent to what you did?

best,
-jacopo

On Mon, May 12, 2014 at 10:41 AM, Peter Steinbach
notifications@github.comwrote:

Hello again,

I ran nvbowtie of these reads:

ftp://ftp-trace.ncbi.nlm.nih.gov/sra/sra-instant/reads/ByExp/sra/SRX%2FSRX298%2FSRX298000/SRR891268/

against the hg-19 genome through this command:

nvBowtie --pe --fr --file-ref hg19-index SRR891268_GM12878_50000cells_1.fastq SRR891268_GM12878_50000cells_2.fastq RR891268.sam

nvBowtie crashes with an error message that is unclear to me:

info : aligning reads [170393600, 170917887]
verbose : 524288 reads
verbose : 52.429 M bps (75.0 MB)
verbose : 50.0 bps/read (min: 50, max: 50)
error : opposite alignment kernel: unspecified launch failure
terminate called after throwing an instance of 'thrust::system::system_error'
what(): unspecified launch failure

Best -

—
Reply to this email directly or view it on GitHubhttps://github.com/NVlabs/nvbio/issues/6#issuecomment-42808079
.

[Deleted user]

Hi Jacopo,

this is Katrin, I am working with Peter.

Indeed, this is what I did too. I renamed the fastq files later to have a better overview of what they represent.

Hope this helps. Best & thanks
Katrin

[jpantaleoni]

Hi Katrin,

thanks a lot - I have been running your dataset against hg19 on a K40, and
couldn't reproduce the issue.

However, this might be due to the fact K40 has more memory than others -
which GPU are you running on?

best,
-jacopo

On Fri, May 16, 2014 at 11:11 AM, ksameith notifications@github.com wrote:

Hi Jacopo,

this is Katrin, I am working with Peter.

Indeed, this is what I did too. I renamed the fastq files later to have a
better overview of what they represent.

Hope this helps. Best & thanks
Katrin

—

Reply to this email directly or view it on GitHubhttps://github.com/NVlabs/nvbio/issues/6#issuecomment-43311859
.

[jpantaleoni]

Nevermind, I can reproduce the error. We will look into it.

On Sat, May 17, 2014 at 11:18 AM, Jacopo Pantaleoni
jpantaleoni@gmail.comwrote:

Hi Katrin,

thanks a lot - I have been running your dataset against hg19 on a K40, and
couldn't reproduce the issue.

However, this might be due to the fact K40 has more memory than others -
which GPU are you running on?

best,
-jacopo

On Fri, May 16, 2014 at 11:11 AM, ksameith notifications@github.comwrote:

Hi Jacopo,

this is Katrin, I am working with Peter.

Indeed, this is what I did too. I renamed the fastq files later to have a
better overview of what they represent.

Hope this helps. Best & thanks
Katrin

—

Reply to this email directly or view it on GitHubhttps://github.com/NVlabs/nvbio/issues/6#issuecomment-43311859
.

[psteinb]

for the record, we are using a K20c and this issue came about with 3e6efbd.
best,
peter