search

synthego-open / ice

Synthego ICE - Inference of CRISPR Editing software
Other
67 stars 29 forks source link

Analysis failed with 311 in the notes #20

Open peilin opened 5 years ago

peilin commented 5 years ago

Hi there,

we use ICE with docker install. It is very easy and fast to use, thanks!

I am curious what exactly the error was if a sample failed analysis. The log shows:

-- analyzing 40 filename.ab1 GGTGAGTGAGTGTGTGCGTG analyzing 462 number of edit proposals R_SQUARED 0.998049444356253 discord (aln window): 0.07 after cutsite: 0.13 Exception Caught! None

I tried on your website, same "311" failed error. Just curious what that means.

Thanks again,

Pei

peilin commented 5 years ago

Hmm, I guess it's tricky to debug this one without the test data. I downloaded the src and ran it locally. This time I got the stack trace:

Exception Caught! Traceback (most recent call last): File "~/ice/ice_env/lib/python3.6/site-packages/ice/analysis.py", line 79, in single_sanger_analysis sa.analyze_sample() File "~/ice/ice_env/lib/python3.6/site-packages/ice/classes/sanger_analysis.py", line 742, in analyze_sample generate_trace_files(self, to_file=trace_file) File "~/ice/ice_env/lib/python3.6/site-packages/ice/outputs/create_trace_files.py", line 117, in generate_trace_files edited_end_bp = sanger_analysis.alignment.ctrl2sample_coords(downstream_in_bp, closest=True) File "~/ice/ice_env/lib/python3.6/site-packages/ice/classes/pair_alignment.py", line 106, in ctrl2sample_coords if self.control_to_sample[coord] is None and closest is True: KeyError: 311

Looks like coord is 311 and not in the dict, I am really new to the sequencing data, any suggestions what may have caused this? Thanks!

peilin commented 5 years ago

I think I know what's going on. Looks like a data quality issue. The experiment ab1 file has 307 reads vs control has 460. It would be good to catch this error upfront and give out a meaningful message like data quality is low.