search

marcus1487 / nanoraw

Genome guided re-segmention and visualization for raw nanopore sequencing data.
https://pypi.python.org/pypi/nanoraw
Other
46 stars 9 forks source link

Start_time not present in fast5 #39

Closed usamec closed 7 years ago

usamec commented 7 years ago

I am using fast5 files from here http://lab.loman.net/2017/03/09/ultrareads-for-nanopore/?rev1 and I get following output:

Failed reads summary: 'start_time' : 42 No events or corrupted events in file. Likely a segmentation error or mis-specified basecall-subgroups (--2d?). : 6

I believe the second error is normal (failed based call). The first error is related to missing start_time attribute in event data and raw data. Can this be fixed somehow? Like setting it to zero?

marcus1487 commented 7 years ago

I have seen this in personal communications with another user, but have yet to find the root of the problem. nanoraw does need the start_time attribute from the fast5 to anchor the events data to the raw data. Without a correct anchoring nanoraw cannot correctly link the event and raw data and thus cannot proceed.

One possibility I did find is that these are from mux scan files. If you save the failed-reads-filename this may point to the issue.

Are you getting some corrected reads, or are all of your reads failing by one of these two reasons?

usamec commented 7 years ago

All files are failing. I think, that we can infer start_time by aligning raw event to called events. I will try at some point and let you know.

marcus1487 commented 7 years ago

I pushed a fix yesterday. If you update to the most recent github you should not see this error anymore.

usamec commented 7 years ago

Works now, thanks.