Strand specificity

[Alexander Predeus]

Hello all, 

I was wondering how important it is for the assembly quality to use the correct strand-specificity settings, and if there's a quick way to check it if you don't have the information? 

Also, on a separate note, is there a way to make logs less clumped, especially remove the progress-bar with the newline? I'm struggling to find the right options. 

Thank you in advance

-- Alex

[Brian Haas]

If you have strand-specific data and you choose the wrong orientation (RF instead of FR, for example), it's fine - you can just reverse complement the resulting transcripts after assembly to get them in the proper orientation.

I'll aim to include some additional helper utilities to explore strand-specificity in the next release.

Your comment about the verbose logging has been noted. :-)

best,

~brian

--
You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-users+unsub...@googlegroups.com.
To post to this group, send email to trinityrnaseq-users@googlegroups.com.
Visit this group at https://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

--

--
Brian J. Haas
The Broad Institute
http://broadinstitute.org/~bhaas

 

[Ken Field]

Alex-

Alternatively, if you told Trinity that your data was stranded when, in fact, it was not then I would recommend repeating the assembly. I think that Trinity would then have assembled each transcript in sense and anti-sense orientation and that would be problematic for downstream analysis.

Lastly, if you told Trinity that your data was not stranded (you left out the flag) when, in fact, it was stranded, then you are probably okay just using the assembly as it is. I think that in that case Trinity would have sorted the reads correctly and assembled the correct orientations for each transcript.

As for a way to check the strand-specificity of an existing dataset, I think that your best bet would be to align it to a similar genome and load the alignment into IGV. Then pick a couple highly expressed genes and look for the orientation of the reads -- it should be very obvious if the library is strand-specific.

Best,

Ken

On Mon, Jan 9, 2017 at 8:00 AM, Brian Haas <bh...@broadinstitute.org> wrote:

If you have strand-specific data and you choose the wrong orientation (RF instead of FR, for example), it's fine - you can just reverse complement the resulting transcripts after assembly to get them in the proper orientation.

I'll aim to include some additional helper utilities to explore strand-specificity in the next release.

Your comment about the verbose logging has been noted. :-)

best,

~brian

On Mon, Jan 9, 2017 at 7:50 AM, Alexander Predeus <pre...@gmail.com> wrote:

Hello all, 

I was wondering how important it is for the assembly quality to use the correct strand-specificity settings, and if there's a quick way to check it if you don't have the information? 

Also, on a separate note, is there a way to make logs less clumped, especially remove the progress-bar with the newline? I'm struggling to find the right options. 

Thank you in advance

-- Alex

--
You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.

To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-users+unsubscribe...@googlegroups.com.

To post to this group, send email to trinityrnaseq-users@googlegroups.com.
Visit this group at https://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

--

--
Brian J. Haas
The Broad Institute
http://broadinstitute.org/~bhaas

 

--
You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-users+unsub...@googlegroups.com.
To post to this group, send email to trinityrnaseq-users@googlegroups.com.
Visit this group at https://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

--

Ken Field, Ph.D.

Professor of Biology

Program in Cell Biology/Biochemistry

Bucknell University

Room 203A Biology Building

[viss...@gmail.com]

Hi Brian

Please excuse my confusion. I have sequencing data from stranded ribo-depleted data generate using a TruSeq stranded mRNA library prep kit (poly-A selection substituted with ribo-depletion) (therefore dUTP sequencing). As I understand from Trinity's manual  I should use RF for assemblies (with read1 as left and read2 as right). However, when I use bowtie2 to map my reads to the reference genome of an ssRNA virus I find that my read (according to botie2's description) is in the forward-reverse (--fr) orientation.

E.g., if --fr is specified and there is a candidate paired-end alignment where mate1 appears upstream of the reverse complement of mate2 and the insert length constraints are met, that alignment is valid. Also, if mate2 appears upstream of the reverse complement of mate1 and all other constraints are met, that too is valid. --rf likewise requires that an upstream mate1 be reverse-complemented and a downstream mate2 be forward-oriented.

It therefore seem like there is a difference between Trinity's RF and bowtie2's --rf. By just reverse-complementing either one of the strands you do not make provision for the orientation of the reads with respect to one-another, in other words whether the one reverse-complement of the one reads is 5' (up-stream) or 3' (down-stream) of the other.

Does Trinity take the orientation of the two reads with respect to each other into account when it performs assemblies? When I specified RF in the downstream applications of Trinity (i.e. differential expression analysis etc.) what parameters did it send to bowtie2, --rf or --fr? And what is the implication thereof for the DE analysis (count matrices)?


I hope you can clarify my confusion the the application of read orientation.
Kind regards
Marike

On Monday, January 9, 2017 at 5:00:30 PM UTC+2, Brian Haas wrote:

If you have strand-specific data and you choose the wrong orientation (RF instead of FR, for example), it's fine - you can just reverse complement the resulting transcripts after assembly to get them in the proper orientation.

I'll aim to include some additional helper utilities to explore strand-specificity in the next release.

Your comment about the verbose logging has been noted. :-)

best,

~brian

On Mon, Jan 9, 2017 at 7:50 AM, Alexander Predeus <pre...@gmail.com> wrote:

Hello all, 

I was wondering how important it is for the assembly quality to use the correct strand-specificity settings, and if there's a quick way to check it if you don't have the information? 

Also, on a separate note, is there a way to make logs less clumped, especially remove the progress-bar with the newline? I'm struggling to find the right options. 

Thank you in advance

-- Alex

--
You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-users+unsub...@googlegroups.com.

To post to this group, send email to trinityrn...@googlegroups.com.

Visit this group at https://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

[Brian Haas]

Hi Marike,

The usage info provides the bowtie or bowtie2 alignment settings used.  In the case of strand-specific options, the strand-specificity flags are set for the various abundance estimation tools, but not at the alignment stage - so all alignments are being captured - the tools are set to use the alignments that make sense given the strand-specific parameterization.

In the case of Trinity, we aim to be consistent on our use of the --SS_lib_type RF setting, though it may differ from other tools as there haven't been standards around this.  Our settings are described here:

https://github.com/trinityrnaseq/trinityrnaseq/wiki/Running%20Trinity#strand_specific_assembly

and we have some additional functionality integrated into our latest release to examine your strand-specific read mappings:

https://github.com/trinityrnaseq/trinityrnaseq/wiki/Examine-Strand-Specificity

I hope this helps,

~brian

To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-users+unsubscribe...@googlegroups.com.

To post to this group, send email to trinityrn...@googlegroups.com.
Visit this group at https://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

--

--
Brian J. Haas
The Broad Institute
http://broadinstitute.org/~bhaas

 

--
You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-users+unsub...@googlegroups.com.

To post to this group, send email to trinityrnaseq-users@googlegroups.com.

Visit this group at https://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

[viss...@gmail.com]

Thanks

To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-users+unsub...@googlegroups.com.

To post to this group, send email to trinityrn...@googlegroups.com.
Visit this group at https://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

--

--
Brian J. Haas
The Broad Institute
http://broadinstitute.org/~bhaas

 

--
You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-users+unsub...@googlegroups.com.

To post to this group, send email to trinityrn...@googlegroups.com.
Visit this group at https://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

[Indugu Nagaraju]

Hello Brian,

My sequence reads are strand-specific.The sequence header are as follows

read_1.fq
@NB500934:8:HLN5GBGX5:1:11101:26019:1048 1:N:0:GAGTGG

read_2.fq
@NB500934:8:HLN5GBGX5:1:11101:26019:1048 2:N:0:GAGTGG

Can I choose the orientation FR?

Please suggest

On Monday, January 9, 2017 at 10:00:30 AM UTC-5, Brian Haas wrote:

If you have strand-specific data and you choose the wrong orientation (RF instead of FR, for example), it's fine - you can just reverse complement the resulting transcripts after assembly to get them in the proper orientation.

I'll aim to include some additional helper utilities to explore strand-specificity in the next release.

Your comment about the verbose logging has been noted. :-)

best,

~brian

On Mon, Jan 9, 2017 at 7:50 AM, Alexander Predeus <pre...@gmail.com> wrote:

Hello all, 

I was wondering how important it is for the assembly quality to use the correct strand-specificity settings, and if there's a quick way to check it if you don't have the information? 

Also, on a separate note, is there a way to make logs less clumped, especially remove the progress-bar with the newline? I'm struggling to find the right options. 

Thank you in advance

-- Alex

--
You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-users+unsub...@googlegroups.com.

To post to this group, send email to trinityrn...@googlegroups.com.

Visit this group at https://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

[Brian Haas]

Hi,

You can't tell the strand-specific orientation from the read names, afaik.  Typically, they're RF, but it depends on the sequencing strategy used.  Here are some tricks to try to figure out what method was used:

https://github.com/trinityrnaseq/trinityrnaseq/wiki/Examine-Strand-Specificity

On Mon, Sep 24, 2018 at 6:49 PM Indugu Nagaraju <ind...@gmail.com> wrote:

Hello Brian,

My sequence reads are strand-specific.The sequence header are as follows

read_1.fq
@NB500934:8:HLN5GBGX5:1:11101:26019:1048 1:N:0:GAGTGG

read_2.fq
@NB500934:8:HLN5GBGX5:1:11101:26019:1048 2:N:0:GAGTGG

Can I choose the orientation FR?

Please suggest


On Monday, January 9, 2017 at 10:00:30 AM UTC-5, Brian Haas wrote:

If you have strand-specific data and you choose the wrong orientation (RF instead of FR, for example), it's fine - you can just reverse complement the resulting transcripts after assembly to get them in the proper orientation.

I'll aim to include some additional helper utilities to explore strand-specificity in the next release.

Your comment about the verbose logging has been noted. :-)

best,

~brian

On Mon, Jan 9, 2017 at 7:50 AM, Alexander Predeus <pre...@gmail.com> wrote:

Hello all, 

I was wondering how important it is for the assembly quality to use the correct strand-specificity settings, and if there's a quick way to check it if you don't have the information? 

Also, on a separate note, is there a way to make logs less clumped, especially remove the progress-bar with the newline? I'm struggling to find the right options. 

Thank you in advance

-- Alex

--
You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.

To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-u...@googlegroups.com.

To post to this group, send email to trinityrn...@googlegroups.com.
Visit this group at https://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

--

--
Brian J. Haas
The Broad Institute
http://broadinstitute.org/~bhaas

 

--

You received this message because you are subscribed to the Google Groups "trinityrnaseq-users" group.

To unsubscribe from this group and stop receiving emails from it, send an email to trinityrnaseq-u...@googlegroups.com.

To post to this group, send email to trinityrn...@googlegroups.com.
Visit this group at https://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.

[raju]

Hello Brian,

Thanks for prompt response.

I am working on the link you have provided. The script 'run_bowtie2.pl' taking too much of time.

For testing can I use subset of reads (i.e ~ 10000 reads).

Is this script supports multi-threading, I tried it but not successful.

Thank you

Best

Raju

  

[Brian Haas]

Right, you can subset your data.  You might need more than 10k reads, though.  Try several million reads and see how that goes.

If something fails, tell us what the error is and we'll have more insights.

best,

~b