Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • Mar 10 15:43

    sahilseth on master

    ci skip: readme update (compare)

  • Mar 10 15:41

    sahilseth on master

    ci skip: badge update (compare)

  • Mar 09 23:19

    sahilseth on master

    ci skip, update badge (compare)

  • Mar 05 22:50

    sahilseth on v0.9.11

    (compare)

  • Mar 02 22:49

    sahilseth on master

    change urls (compare)

  • Mar 02 20:16

    sahilseth on master

    change path encoding for R 3.3/… (compare)

  • Mar 02 20:10

    sahilseth on master

    updating all URLs in docs and v… (compare)

  • Feb 28 05:00

    sahilseth on master

    add params (compare)

  • Feb 28 04:56

    sahilseth on master

    bump params version (compare)

  • Feb 28 04:35

    sahilseth on master

    adding news, fixing issue with … Merge pull request #101 from fl… (compare)

  • Feb 28 04:35
    sahilseth closed #101
  • Feb 28 04:35
    sahilseth opened #101
  • Feb 28 04:25

    sahilseth on devel

    adding news, fixing issue with … (compare)

  • Feb 28 04:11

    sahilseth on master

    add a new function to get detai… Merge branch 'devel' of https:/… bump version, update flowr url and 1 more (compare)

  • Feb 28 04:11
    sahilseth closed #100
  • Feb 28 02:59
    sahilseth synchronize #100
  • Feb 28 02:59

    sahilseth on devel

    bump version, update flowr url (compare)

  • Feb 15 16:35
    sahilseth added as member
  • Feb 15 16:32

    sahilseth on gh-pages

    Delete CNAME (compare)

  • Feb 09 17:17

    sahilseth on master

    skip ci (compare)

Samir B. Amin
@dyndna
Do you have yesterday's final running flowR RData file?
Samir B. Amin
@dyndna
minor bug: ```> plot_flow(fobj)
Error in (function (classes, fdef, mtable) :
unable to find an inherited method for function ‘plot_flow’ for signature ‘"list"’

and major one:-) ```

submit_flow(fobj)
Error in checkAtAssignment("list", "execute", "logical") :
‘execute’ is not a slot in class “list”
```

Samir B. Amin
@dyndna
minor one if applied to flowr: dir.exists function is not present in R < 3.2. In case if you are using it widely.
sseth
@sahilseth
Submit_flow and plot_flow used to work with a single flow object.
Submit_flow now works with a list of flow objects (tested). plot_flow also works, but am not sure if a user would like one panel with all the plots, or would like R to make all of them one by them. What if we have a hundred flow objects in the list. Ideally, if fobjs is a list, it is best to use: plot_flow(fobs[[1]]) and plot one of them, since all the plots would be identical.
sseth
@sahilseth
@dyndna I don't use dir.exists. Additionally our cluster is still on 3.1.3, so all the code is being tested on that only.
sseth
@sahilseth
Tommy asked:
mtang1@cnode338:~/flowr/runs$ flowr status x=`pwd`

================================================================================
Summarizing status (using triggers) of: 
/scratch/genomic_med/mtang1/flowr/runs
  |======================================================================| 100%
|        | total| started| completed| exit_status|status     |
|:-------|-----:|-------:|---------:|-----------:|:----------|
|001.rln |    12|      12|         8|           0|processing |
|002.snv |     6|       4|         4|           0|processing |
|003.stv |     6|       4|         4|           1|errored    |
  1. we need to find the run which failed:
flowr status x=/scratch/genomic_med/mtang1/flowr/runs/*
Say the run is ~/flowr/runs/my_aweome_run-2015..... and it failed at step stv.
We would do
flowr rerun x=~/flowr/runs/my_aweome_run-2015.... start_from=stv
Ming Tang
@crazyhottommy
got it
thx
what if I deleted the run with errors in the run folder?.... and also the report folder...
sseth
@sahilseth
Do you know what the error is ? Re-run is useful so that step rln and stv would not be repeated. If you delete you need to start that run from scatch
Ming Tang
@crazyhottommy
Sorry, I was in a meeting with Ayush.
Ming Tang
@crazyhottommy
actually, the walltime is not enough...I deleted them...but when I rerun it from the very beginning. it complains:
Ming Tang
@crazyhottommy
Warning message:
In file.rename("Rplots.pdf", file.path(myoutdir, sprintf("flowr_%s_plots_%s.pdf",  :
  cannot rename file 'Rplots.pdf' to '../reports/LGG/flowr_LGG_plots_X16_Oct_15_081224CDT.pdf', reason 'No such file or directory'

I guess, flow still remembers the old file flowr_LGG_plots_X16_Oct_15_081224CDT.pdf

Today is already Oct_16

sseth
@sahilseth
Sorry was no clear, this warning is not from flowr, but from the wrapper script Samir created.
If wall time is a problem, and only the last step failed. All you need to do is:
We need a new updated flow definition with more wall time
then do,
flowr rerun  x=~/flowr/runs/my_aweome_run-2015.... start_from=stv def=my_new_def.def
Ming Tang
@crazyhottommy
gotcha! thx
Ming Tang
@crazyhottommy
@sahilseth the speedseq pipeline by samir has three steps. 1. realign 2. snv 3. stv .
2 and 3 depends on 1.
2 and 3 can be run in parallel.
If I only want to run 1 and 3, how can I do it? Thanks!
sseth
@sahilseth
I am not sure how to do that. Are you using the code from gitlab?
Ming Tang
@crazyhottommy
Ming Tang
@crazyhottommy
@sahilseth the speedseq pipeline by samir has three steps. 1. realign 2. snv 3. stv .
2 and 3 depends on 1.
2 and 3 can be run in parallel.
when I do flowr rerun x=path_to_my_flow start_from=snv flowr will rerun both snv and stv. Is this expected? I only want to rerun snv in this case. Thanks
Samir B. Amin
@dyndna
I think this is expected as stv step follows snv. Perhaps, flowr should have "end_at" like argument to specify range of intermediate steps to run. It may have it already, I've not checked it.