reference_classif.txt

[pawel-wojciechowski]

Hi,
does anyone know what "The path to the reference transcriptome classification file is /path/to/reference_classif.txt" means . Which file is it exactly? It appears as a parameter of the SQANTI3 rescue step.

Pawel

[carolinamonzo]

Hi @pawel-wojciechowski it's explained in the SQANTI wiki / Running SQANTI3 rescue / SQANTI3 resque strategy / 3. Application of SQ3 filter to the reference transcriptome. You need to generate the reference_classif.txt file by running SQANTI3 QC on your reference transcriptome, using also the same orthogonal data you used on your samples.

I hope this solves your question!
Best,
Carolina.

[pawel-wojciechowski]

Hi Caroline,
indeed everything is described. Sorry for the confusion. I would recommend possibly adding a link to the Wiki in the https://github.com/ConesaLab/SQANTI3/wiki/Tutorial:-running-SQANTI3-on-an-example-dataset#Executing-and-Understanding-SQANTI3-Rescue section to avoid further questions (it's not obvious in the video tutorial). I don't know why but somehow I found it to be the most accurate description.
In any case, thank you!

Great job!
Best regards,
Pawel.

[mpizzagalli777]

Sorry I have a quick follow up question to this. When running the reference transcriptome, is it simply as follows? I was a little unclear on how to run QC on the transcriptome correctly.

python $sqanti/sqanti3_qc.py $genome/gencode.v44.annotation.gtf $genome/gencode.v44.annotation.gtf $genome/GRCh38.p14.genome.fa \
	-c $sr_dir/G2_1_filtered_SJ.out.tab,$sr_dir/G2_2_filtered_SJ.out.tab \
	--SR_bam $wkdir/bam_list.txt \
	-o gencode.v44.annotation -d SQANTI3_output --report skip --force_id_ignore

Thanks for the help!

[carolinamonzo]

Hi, yes, exactly, you simply run it with the reference genome GTF as input (like as if you were running the GTF of a sample of interest). After this, when you run sqanti3_resque, remember to add the output *_classification.txt to --refClassif.

[MR-D-CJ]

To generate the reference_classif.txt file, when I run sqanti3_qc.py against reference, my input data is the reference transcript data, something like the following, is this correct?

python sqanti3_qc.py \ /PATH/GRCh38/genecode/gencode.v47.transcripts.fa \ /PATH/GRCh38/genecode/gencode.v46.annotation.gtf \ /PATH/GRCh38/GCA_000001405.15_GRCh38_no_alt/GCA_000001405.15_GRCh38_no_alt_analysis_set.fa \ --CAGE_peak /PATH/human.refTSS_v3.1.hg38.sorted.bed \ --polyA_motif_list /PATH/polyA_info/mouse_and_human.polyA_motif.txt \ -o GRCh38 \ -d /PATH/SQANTi3_ref/GRCh38_QC \ --fasta \ --force_id_ignore \ --cpus 20 --report html

[carolinamonzo]

Hi, yes, sorry, I forgot to answer

[MR-D-CJ]

Hi, yes, sorry, I forgot to answer

Thank u！

[MR-D-CJ]

sorry,I have another problem.I run sqanti.qc for collapsed file. when i use refTSS_v4.1_human_coordinate.hg38.bed.txt from https://reftss.riken.jp/datafiles/4.1/human/,
can't create *_classification.txt, i don't know why?
code is
python /share/home/duchangjian/soft/biosoft/SQANTI3-5.2.2/sqanti3_qc.py \ /PATH/Sample.GRCh38-no-alt.collapsed.gff \ /PATH/human/GRCh38/genecode/gencode.v46.annotation.gtf \ /PATH/human/GRCh38/GCA_000001405.15_GRCh38_no_alt/GCA_000001405.15_GRCh38_no_alt_analysis_set.fa \ --CAGE_peak /PATH/refTSS/human.refTSS_v4.1.hg38.sorted.bed \ --polyA_motif_list /PATH/polyA_info/mouse_and_human.polyA_motif.txt \ -o SampleGRCh38 \ -d /PATH/sqanti.refTSS4 \ -fl /PATH/Sample.GRCh38-no-alt.collapsed.abundance.txt \ --cpus 20 --report html 2>>"$LOGFILE"

ouptput file only have:

log file show that:

refTSS_v4.1_human_coordinate.hg38.bed.txt file is