z0on
commented
5 years ago Hi Devin - With Trinity you don’t need steps 2 and 3, they are only for cases when you don’t have assembler-derived genes. Was the file “transcriptome_seq2iso.tab” non-empty? looking reasonable? Can you paste the first 10 lines? Mish
On Sep 10, 2019, at 4:20 PM, LibraLizard66 notifications@github.com wrote:
Hi all,
I'm trying to assay differential gene expression from 22 RNA samples, two of them sequenced with RNAseq and 20 of them sequenced with TagSeq. I used the RNAseq sequences to generate a reference transcriptome de novo via Trinity to compare the TagSeq sequences to.
I've been following the pipeline listed here: https://github.com/z0on/tag-based_RNAseq/blob/master/tagSeq_processing_README.txt https://github.com/z0on/tag-based_RNAseq/blob/master/tagSeq_processing_README.txt Everything went smoothly until the "# generating read-counts-per gene" section. Here's a rundown of what I did and where things appeared wrong (Skip to Step 4 if you want to go straight to that).
STEP 1 I used the command: grep ">" Trinity.fasta | perl -pe 's/^>(\S+)|(\S+)(i\d+)\s.+/$1|$2$3\t$1$2/' >transcriptome_seq2iso.tab in order to create a two-column tab-delimited table to give correspondence between entries in the transcriptome fasta file and genes (our transcriptome fasta file was named "Trinity.fasta", not "trinity.fasta" as per the readme file). The file "transcriptome_seq2iso.tab" was created.
STEP 2 The pipeline afterwards lists some steps to take if you don't have any assembler-derived genes (cd-hit-est). As we had plenty of assembler-derived-genes, I planned on skipping it, but isogroup_namer.pl (in the next step) seems to depend on having ch-hit-est. Thus, I ran cd-hit-est using the following code: cdhit-est -i Trinity.fasta -o transcriptome_clust.fasta -c 0.99 -G 0 -aL 0.3 -aS 0.3 The following two files were created 1) transcriptome_clust.fasta and 2) transcriptome_clust.fasta.clustr
STEP 3 To try to add cluster designations to the fasta headers, I used the code: /tag-based_RNAseq/isogroup_namer.pl transcriptome_clust.fasta transcriptome_clust.fasta.clstr >transcriptome_seq2gene.tab Note: The "/tag-based_RNAseq" portion of the code was to tell Ubuntu where isogroup_namer.pl was. The following four files were created: 1) transcriptome_clust_iso.fasta 2) transcriptome_clust_iso.seq2iso.tab 3) transcriptome_seq2gene.tab and 4) transcriptome_seq2iso.tab However, when I examined these files with nano, all of the files were empty except for transcriptome_seq2iso.tab (the 4th one), which, as far as I could tell, looked like it should.
STEP 4 (THIS IS WHERE THINGS HAVE APPARENTLY GONE WRONG) To try to count the hits per isogroup, I ran the following code: /home/genomics/tag-based_RNAseq/samcount_launch_bt2.pl '.sam' /Data/devin/JA19193/transcriptome_seq2iso.tab > sc Note: "/home/genomics/tag-based_RNAseq" pointed Ubuntu to samcount_launch_bt2.pl, and "/Data/devin?JA19193" pointed Ubuntu towards transcriptome_seq2.iso.tab. After doing this, I used nano to edit sc to add the full path to samcount before each sample so Ubuntu could find samcount. For example: samcount.pl 10_merged.trim.sam /home/genomics/Data/devin/JA19193/transcriptome_seq2iso.tab aligner=bowtie2 >10_merged.trim.sam.counts became ~/tag-based_RNAseq/samcount.pl 10_merged.trim.sam /home/genomics/Data/devin/JA19193/transcriptome_seq2iso.tab aligner=bowtie2 >10_merged.trim.sam.counts Note: "10_merged" is the name of a TagSeq sample.
I then executed sc using the following code: bash sc
samcount ran, but in every line of output while it ran, the message, "[gene name] has no isogroup designation” appeared. I tried creating the sc file using the file "transcriptome_seq2gene.tab" (the 3rd file in Step 3 above) and running samcount with this modified sc, but received the same result. Thus, it appears that samcount is treating the only meaningful product file from isogroup_namer as an empty file. Can anyone give me some help? I'm running Ubuntu version 16.04.
Thanks, -Devin
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/z0on/tag-based_RNAseq/issues/1?email_source=notifications&email_token=ABZUHGDQGMRQDNWO4ECE65LQJAFRHA5CNFSM4IVMWOB2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HKR3WTQ, or mute the thread https://github.com/notifications/unsubscribe-auth/ABZUHGCRTPOSD66W6DS677LQJAFRHANCNFSM4IVMWOBQ.
LibraLizard66
paigeduffin
Hi all,
I'm trying to assay differential gene expression from 22 RNA samples, two of them sequenced with RNAseq and 20 of them sequenced with TagSeq. I used the RNAseq sequences to generate a reference transcriptome de novo via Trinity to compare the TagSeq sequences to.
I've been following the pipeline listed here: https://github.com/z0on/tag-based_RNAseq/blob/master/tagSeq_processing_README.txt Everything went smoothly until the "# generating read-counts-per gene" section. Here's a rundown of what I did and where things appeared wrong (Skip to Step 4 if you want to go straight to that).
STEP 1 I used the command: grep ">" Trinity.fasta | perl -pe 's/^>(\S+)|(\S+)(i\d+)\s.+/$1|$2$3\t$1$2/' >transcriptome_seq2iso.tab in order to create a two-column tab-delimited table to give correspondence between entries in the transcriptome fasta file and genes (our transcriptome fasta file was named "Trinity.fasta", not "trinity.fasta" as per the readme file). The file "transcriptome_seq2iso.tab" was created.
STEP 2 The pipeline afterwards lists some steps to take if you don't have any assembler-derived genes (cd-hit-est). As we had plenty of assembler-derived-genes, I planned on skipping it, but isogroup_namer.pl (in the next step) seems to depend on having ch-hit-est. Thus, I ran cd-hit-est using the following code: cdhit-est -i Trinity.fasta -o transcriptome_clust.fasta -c 0.99 -G 0 -aL 0.3 -aS 0.3 The following two files were created 1) transcriptome_clust.fasta and 2) transcriptome_clust.fasta.clustr
STEP 3 To try to add cluster designations to the fasta headers, I used the code: ~/tag-based_RNAseq/isogroup_namer.pl transcriptome_clust.fasta transcriptome_clust.fasta.clstr >transcriptome_seq2gene.tab Note: The "~/tag-based_RNAseq" portion of the code was to tell Ubuntu where isogroup_namer.pl was. The following three files were created: 1) transcriptome_clust_iso.fasta 2) transcriptome_clust_iso.seq2iso.tab 3) transcriptome_seq2gene.tab However, when I examined these files with nano, all of the files were empty.
STEP 4 (THIS IS WHERE THINGS HAVE APPARENTLY GONE WRONG) To try to count the hits per isogroup, I ran the following code: /home/genomics/tag-based_RNAseq/samcount_launch_bt2.pl '.sam' ~/Data/devin/JA19193/transcriptome_seq2iso.tab > sc Note: "/home/genomics/tag-based_RNAseq" pointed Ubuntu to samcount_launch_bt2.pl, and "~/Data/devin?JA19193" pointed Ubuntu towards transcriptome_seq2.iso.tab. After doing this, I used nano to edit sc to add the full path to samcount before each sample so Ubuntu could find samcount. For example: samcount.pl 10_merged.trim.sam /home/genomics/Data/devin/JA19193/transcriptome_seq2iso.tab aligner=bowtie2 >10_merged.trim.sam.counts became ~/tag-based_RNAseq/samcount.pl 10_merged.trim.sam /home/genomics/Data/devin/JA19193/transcriptome_seq2iso.tab aligner=bowtie2 >10_merged.trim.sam.counts Note: "10_merged" is the name of a TagSeq sample.
I then executed sc using the following code: bash sc
samcount ran, but in every line of output while it ran, the message, "[gene name] has no isogroup designation” appeared. I tried creating the sc file using the file "transcriptome_seq2gene.tab" (the 3rd file in Step 3 above) and running samcount with this modified sc, but received the same result. Thus, it appears that samcount is treating the only meaningful product file from isogroup_namer as an empty file. Can anyone give me some help? I'm running Ubuntu version 16.04.
Thanks, -Devin