Calculating Expressed Transcripts from Multiple Samples
84 views
Skip to first unread message
joe.cor...@gmail.com
Nov 5, 2015, 12:05:12 PM11/5/15
to trinityrnaseq-users
The transcript abundance estimation method outlined in the Trinity documentation states that estimated counts should be generated for each sample. With enough samples it becomes cumbersome to use individual abundance estimations for judging the quality of an assembly.
Is it possible to combine the counts to generate single ExN50/E90 stats from multiple samples of transcript count estimates?
Is it possible to combine the counts to generate single ExN50/E90 stats from multiple samples of transcript count estimates?
Brian Haas
Nov 5, 2015, 12:27:15 PM11/5/15
to joe.cor...@gmail.com, trinityrnaseq-users
The script is supposed to take the full expression matrix as input, so it uses all samples.
best,
~b
On Thu, Nov 5, 2015 at 12:05 PM, <joe.cor...@gmail.com> wrote:
The transcript abundance estimation method outlined in the Trinity documentation states that estimated counts should be generated for each sample. With enough samples it becomes cumbersome to use individual abundance estimations for judging the quality of an assembly.
Is it possible to combine the counts to generate single ExN50/E90 stats from multiple samples of transcript count estimates?
--
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 http://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.
joe.cor...@gmail.com
Nov 5, 2015, 1:19:36 PM11/5/15
to trinityrnaseq-users, joe.cor...@gmail.com
Very sorry, misread the example there as taking pair-wise samples.
On Thursday, November 5, 2015 at 12:27:15 PM UTC-5, Brian Haas wrote:
The script is supposed to take the full expression matrix as input, so it uses all samples.best,~b
On Thu, Nov 5, 2015 at 12:05 PM, <joe.cor...@gmail.com> wrote:
The transcript abundance estimation method outlined in the Trinity documentation states that estimated counts should be generated for each sample. With enough samples it becomes cumbersome to use individual abundance estimations for judging the quality of an assembly.
Is it possible to combine the counts to generate single ExN50/E90 stats from multiple samples of transcript count estimates?
--
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 http://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.
Brian Haas
Nov 5, 2015, 1:22:32 PM11/5/15
to joe.cor...@gmail.com, trinityrnaseq-users
no worries! Best of luck! (looking forward to hearing how good your ExN50 is! ;)
~b
On Thu, Nov 5, 2015 at 1:19 PM, <joe.cor...@gmail.com> wrote:
Very sorry, misread the example there as taking pair-wise samples.
On Thursday, November 5, 2015 at 12:27:15 PM UTC-5, Brian Haas wrote:
The script is supposed to take the full expression matrix as input, so it uses all samples.best,~b
On Thu, Nov 5, 2015 at 12:05 PM, <joe.cor...@gmail.com> wrote:
The transcript abundance estimation method outlined in the Trinity documentation states that estimated counts should be generated for each sample. With enough samples it becomes cumbersome to use individual abundance estimations for judging the quality of an assembly.
Is it possible to combine the counts to generate single ExN50/E90 stats from multiple samples of transcript count estimates?
--
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 http://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.
--
--
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 http://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.
joe.cor...@gmail.com
Nov 5, 2015, 2:19:13 PM11/5/15
to trinityrnaseq-users, joe.cor...@gmail.com
I'm currently getting the following error from abundance_estimates_to_matrix.pl:
Use of uninitialized value within %fields in array element at /opt/trinityrnaseq/util/abundance_estimates_to_matrix.pl line 164, <$fh> line <XXX>.
It is being thrown seemingly constantly when trying to run on the abundance.tsv files generated in the previous step (align_and_estimate_abundance.pl).
During the align_and_estimate phase, I did see that warnings stating that there weren't enough fragments to estimate the bias parameters, is this a result of that?
Use of uninitialized value within %fields in array element at /opt/trinityrnaseq/util/abundance_estimates_to_matrix.pl line 164, <$fh> line <XXX>.
It is being thrown seemingly constantly when trying to run on the abundance.tsv files generated in the previous step (align_and_estimate_abundance.pl).
During the align_and_estimate phase, I did see that warnings stating that there weren't enough fragments to estimate the bias parameters, is this a result of that?
On Thursday, November 5, 2015 at 1:22:32 PM UTC-5, Brian Haas wrote:
no worries! Best of luck! (looking forward to hearing how good your ExN50 is! ;)~b
On Thu, Nov 5, 2015 at 1:19 PM, <joe.cor...@gmail.com> wrote:
Very sorry, misread the example there as taking pair-wise samples.
On Thursday, November 5, 2015 at 12:27:15 PM UTC-5, Brian Haas wrote:
The script is supposed to take the full expression matrix as input, so it uses all samples.best,~b
On Thu, Nov 5, 2015 at 12:05 PM, <joe.cor...@gmail.com> wrote:
The transcript abundance estimation method outlined in the Trinity documentation states that estimated counts should be generated for each sample. With enough samples it becomes cumbersome to use individual abundance estimations for judging the quality of an assembly.
Is it possible to combine the counts to generate single ExN50/E90 stats from multiple samples of transcript count estimates?
--
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 http://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.
--
--
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 http://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.
Brian Haas
Nov 5, 2015, 2:30:32 PM11/5/15
to joe.cor...@gmail.com, trinityrnaseq-users
Hi Joe,
What's the command you ran?
~b
Brian Haas
Nov 5, 2015, 2:55:16 PM11/5/15
to Joe Cornish, trinityrn...@googlegroups.com
Ah! If you're using the very latest version of Trinity, it'll be compatible with v1.5 of eXpress. If you're using an older version of Trinity, you might need an older version of eXpress.
If all else fails, try using Kallisto or RSEM, as they're also both supported in the latest release.
best,
~b
On Thu, Nov 5, 2015 at 2:46 PM, Joe Cornish <joe.cor...@gmail.com> wrote:
$outdir/<sample id>/abundance.tsvPart of a script:the abundance.tsv files are in a dirs with the pattern
#abundance dir
abundance_estimates_to_matrix.pl \
--est_method eXpress \
--out_prefix trans \
--name_sample_by_basedir \
`ls $outdir/*/abundance.tsv` 2> "$outdir/mat.log"I still get a final output from contig_ExN50_statistic.pl
joe.cor...@gmail.com
Nov 5, 2015, 3:00:40 PM11/5/15
to trinityrnaseq-users, joe.cor...@gmail.com
Probably won't be able to get the admin to install the new version till next week, any chance I can just grab the new version of abundance_estimates_to_matrix.pl, or is there more to it than that?
Brian Haas
Nov 5, 2015, 3:02:22 PM11/5/15
to Joe Cornish, trinityrnaseq-users
Sure. Try the latest version and see if that works:
--
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 http://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.
joe.cor...@gmail.com
Nov 5, 2015, 3:12:32 PM11/5/15
to trinityrnaseq-users, joe.cor...@gmail.com
Nope, same problem, I guess I'll have to wait. 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 http://groups.google.com/group/trinityrnaseq-users.
For more options, visit https://groups.google.com/d/optout.
Reply all
Reply to author
Forward
0 new messages