Simulator fails using different reads about nanosim

[JustinChu]

Hi,

Luckily, I was able to reproduce the error quickly because I actually already had this dataset on hand. Thanks for letting us know. We will try to have a fix soon ASAP.

I fixed one bug related to this (a simple parsing error) but there still seems to be a bug related to the alignment ratio. The reads are indeed much better (longer and higher quality) than before. 100% of the reads map and the average reads lengths are much longer. This is causing problems in the parsing of the MAP_align_ratio file created.

@cheny19 I'm now convinced that our pre-set bins sizes for read lengths ranges are not a good idea. All the reads are in the align_ratio file are migrating to the last bin (because the reads are so long), leaving every other bin empty, cauinge errors when loading this file. I can change the bins to scale to the data but I don't know what effect this will have on the modeling.

[cheny19]

Hi Mike,

Sorry I didn't get back to you earlier as I was on vacation.
The binning problem is fixed now. I'm using equal number of reads in each bin now, and user can specify the number of bins they need. The default number of bins is 20.
I still need to test and find out the best default value that works for most of the datasets, but right now, this version is ready to use. I tested it on your dataset and it works pretty well.
Let me know if you have further concerns.

[ml12220]

That works fine now thanks!