how I could know whether my data sets coming from RNA-seq are strand-specific or not????? I emailed the company but they are in Chinese new year and could not help me yet but I should start the analysis.
Align a million or so reads (using the unstranded mode, if your aligner has one) and then use RSeQC to infer the experiment type.
Just to complete your suggestion with a link =)
thank you, then I need both bam and bed files? but in UCSC Table Browser there is no such a bed for Arabidopsis, then what to do?
Get the bed from BioMart (Ensembl plant): http://plants.ensembl.org/Arabidopsis_thaliana/Info/Index
Make sure the genome build matches.
thank you, for aligning I used gtf coming from ensembl but there is gff3. no problem???
the main problem is this, I tried to convert my gtf to bed by bedops but some bug that I could not solve then I tried galaxy but in galaxy there is only gff to bed then I used my gtf instead of gff but
[izadi@lbox161 bin]$ infer_experiment.py -r Galaxy142-[GFF-to-BED_on_data_141].bed -i 204.bam
infer_experiment.py: No match.
then do you know another solution please?
./infer_experiment.py -r Galaxy142-[GFF-to-BED_on_data_141].bed -i 204.bam
Looks like you need to provide the path for infer_experiment.py since that does not seem to be in that bin directory.
thank you but please consider
[izadi@lbox161 bin]$ pwd
[izadi@lbox161 bin]$ ls
Try this in case Devon's suggestion did not work
$ python ./infer_experiment.py -r Galaxy142-[GFF-to-BED_on_data_141].bed -i 204.bam
[izadi@lbox161 bin]$ python infer_experiment.py -r Galaxy142-[GFF-to-BED_on_data_141].bed -i 204.bam
python: No match.
You don't have python installed on this machine? What OS are you using?
$ which python
[izadi@lbox161 bin]$ which python
Looks like python/infer_experiment.py does not like that atrocious BED file name. Can you move/copy it to something else and try.
$ cp Galaxy142-[GFF-to-BED_on_data_141].bed simple.bed
$ ./infer_experiment.py -r simple.bed -i 204.bam
[izadi@lbox161 bin]$ infer_experiment.py -r simple.bed -i 204.bam
Traceback (most recent call last):
File "infer_experiment.py", line 49, in <module>
from bx.bitset import *
ImportError: No module named bx.bitset
[izadi@lbox161 RSeQC-2.6.3]$ python setup.py build
[izadi@lbox161 RSeQC-2.6.3]$ sudo python setup.py install
[sudo] password for izadi:
Sorry, user izadi is not allowed to execute '/bin/python setup.py install' as root on lbox161.mpikg.mpg.de.
python setup.py install --root=/home/user/XXX/
You could ask your sys admins to install the modules or try to install them in your own directory: https://www.seas.upenn.edu/cets/answers/install-python-module.html
It is a bit late in this thread but can't you ask whoever produced this data if they used a "strand-specific" protocol?
I emailed the company but they are in holliday
Not ideal but you could look at the alignments in IGV/Tablet to see if you can see an identifiable pattern in the alignment of reads.
thank you so much
in a biostars post I read something about sam flags
then i saw
abundances of 83, 99, 147 and 163 flags of my sam file is the same=1 , then might library IS sequenced strand-specific
anyway I think they will back from holiday Feb 14 and may they reply my email
sorry, I installed Python then this is my result
Reading reference gene model simple.bed ... Done
Loading SAM/BAM file ... Total 200000 usable reads were sampled
This is PairEnd Data
Fraction of reads failed to determine: 0.0068
Fraction of reads explained by "1++,1--,2+-,2-+": 0.4992
Fraction of reads explained by "1+-,1-+,2++,2--": 0.4940
my library was sequenced in strand-specific mode? sorry if so first or second strand?
No it was not strand-specific. See the explanation here: http://rseqc.sourceforge.net/#infer-experiment-py
Eeek, I had a few missing words in there too! Thanks for adding the link!
I aligned but totally on transcriptom and I have a accepted_hits.bam now.
now I am in this path
then how I can detect if is strand-specific?