Jim Robinson

@rbuels You are surely among the handful of people who know the details of CRAM, but I imagine that would be very complex. Perhaps it would solve the out-of-memory error,...

Hi all, I might have a solution for this based on jsfive here: https://github.com/jrobinso/hdf5-indexed-reader. We are now remotely accessing individual datasets from HDF5 file ~ 180 GB in size in...

Just an opinion, but I think you will find more tools that expect BED files to be tab-delimited than whitespace delimited. I don't think UCSC ever defined this. I don't...

Ahh, I was wrong, UCSC does provide some guidance here, note step 1. `Step 1. Format the data set: Format your data as a tab-separated file...` https://genome-preview.gi.ucsc.edu/goldenPath/help/customTrack.html

@ArtRand apologies for that, I am the IGV developer, that is just a bug. You can raise an issue at https://github.com/igvteam/igv or (for the web) at https://github.com/igvteam/igv.js when problems arise,...

BTW bigbed is recommended for files this size anyway (my test file is 125 MB). If making a bigbed file use the autosql option to describe your columns.

Also, I can't reproduce an error with my test data in IGV desktop. I do have to rename the extension to ".bed", the ".bedmethyl" extension will be added to the...

@ArtRand When parsing "bed" files IGV continues to read columns until it encounters something unexpected. For example a non-numeric string when expecting a number, etc. This isn't perfect but works...

@ArtRand thanks, I see why this one failed and my other test file did not. The extra columns actually look o.k. in the working test file, they are being parsed...

Yes, I was able to code IGV to support any type of delimiter for columns beyond the standard fields. This is possible because we added a specific type, i.e. it...