Error in log file on completed run

[AMChalkie]

Hi,

A couple of questions:

• Do any dep files exist for mouse / how to generate these?
• I see Execution halted, but completed results files. Whats going on there?

Best wishes
Alistair

• docker run --tty -u $(id -u):$(getent group svi-stemcell | cut -d: -f3) --mount type=bind,source=/mnts/indicies,target=/mnts/indicies --mount type=bind,source=/mnts/scratch,target=/mnts/scratch --mount type=bind,source=/mnts/bioinfo,target=/mnts/bioinfo ndatth/circall:v1.0.1 Circall.sh
  -genome /mnts/scratch/circrna_using_Circall_docker/data/Mus_musculus.GRCm38.dna.primary_assembly.fa
  -gtfSqlite /mnts/scratch/circrna_using_Circall_docker/data/Mus_musculus.GRCm38.102.sqlite
  -txFasta /mnts/scratch/circrna_using_Circall_docker/data/Mus_musculus.GRCm38.cdna.all.fa
  -txIdx /mnts/scratch/circrna_using_Circall_docker/data/IndexTranscriptome
  -bsjIdx /mnts/scratch/circrna_using_Circall_docker/data/IndexBSJ
  -read1 /mnts/scratch/sgMDA5_total_d3_rep3_CW_18_1.fq.gz
  -read2 /mnts/scratch/sgMDA5_total_d3_rep3_CW_18_2.fq.gz
  -p 12
  -tag sgMDA5_total_d3_rep3
  -c FALSE
  -td FASLE
  -o /mnts/scratch/circrna_using_Circall_docker/output/sgMDA5_total_d3_rep3
  )

            Do not compute fdr2d, just export to file!!

---

                 _
        _____   (_)   _____  _____   __        __      __
       / ___/  / /   / ___/ / ___/  /  \      / /     / /
      / /__   / /   / /    / /__   / __ \    / /__   / /__
      \___/  /_/   /_/     \___/  /_/  \_\  /_/__/  /_/__/

---

                        DONE!

---

tail /mnts/scratch/circrna_using_Circall_docker/output/sgMDA5_total_d3_rep3/sgMDA5_total_d3_rep3_Circall.log
3286 1 0.1255279 292.0
3289 1 0.1255279 1109.0
3298 1 0.1255279 742.0
3340 1 0.1255279 494.0
3346 1 0.1255279 518.0
3483 1 0.1255279 764.0
3539 1 0.1255279 1927.0

Done! Results are save in /mnts/scratch/circrna_using_Circall_docker/output/sgMDA5_total_d3_rep3/sgMDA5_total_d3_rep3_Circall_final.txtError: object 'PEres' not found
Execution halted

head /mnts/scratch/circrna_using_Circall_docker/output/sgMDA5_total_d3_rep3/sgMDA5_total_d3_rep3_Circall_final.txt
chr start end geneID circID junction_fragment_count junction_FPM median_circlen
6 40683939 40709120 ENSMUSG00000068587 6__40683939__40709120 5 0.62763969564998 2685
8 33669153 33685702 ENSMUSG00000031584 8__33669153__33685702 5 0.62763969564998 735

[datngu]

Hi,
Thank you very much for using the tool and reporting the bugs to us!

I will investigate the error this weekend. It would be great if you can provide me some ribo-minus RNA seq data in mouse (any - public data is good so that I can download it) so I can test directly on the mouse genome and annotations.

Regarding the dep data, I put one file with fdr2d parameters estimated from human datasets in the docker image. You can use this option to enable it:
-dep Circall/Data/Circall_depdata_human.RData

• Regarding generating the dep file for the mouse genome, it typically needs both ribo-minus RNA seq and its corresponding ribo-minus RNA-R enzyme treated RNA seq (this enzyme aims to remove linear RNA so we would yield the true positive - this assumption may not be perfect but is common for circRNA detection) circRNA). By comparing the 2 library results, fdr2d estimates parameters and thus can prioritize "true" circRNAs.

• we also have the script to getfdr2d https://github.com/datngu/Circall/blob/master/Circall_v1.0.1/R/getFdr.R but it seems to be a bug in docker, I will revisit that and update to you!

Hope that this is useful!

Best wishes and have a nice weekend!
Dat

[AMChalkie]

Thanks :) I appreciate the effort.

Here is an example dataset but I'm not 100% sure it is what you need.

https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE83823

Best wishes
Alistair