UCSC's NCBI RefSeq Track tables: header differences

Question

[Synanth]

Hi,

I'm working with a piece of software that requires RefSeq track tables, and I'm running into issues when trying to update from hg38 to chm13. The following are the headers for each table:

hg38: bin name chrom strand txStart txEnd cdsStart cdsEnd exonCount exonStarts exonEnds score name2 cdsStartStat cdsEndStat exonFrames

chm13: chrom chromStart chromEnd name score strand thickStart thickEnd reserved blockCount blockSizes chromStarts name2 cdsStartStat cdsEndStat exonFrames type geneName geneName2 geneType

Is there a way to translate the chm13 file to have the same format as hg38 (perhaps involving the bb file)? Or am I SOL in that there is no translation.

Thank you

Answer 1

[Pierre Lindenbaum]

My guess is that chromStarts plus blockSize is equivalent to exonStarts/exonEnds. If you need the bin column, you'll have to script things; see https://genomewiki.ucsc.edu/index.php/Bin_indexing_system / https://github.com/lh3/misc/blob/cc0f36a9a19f35765efb9387389d9f3a6756f08f/biodb/batchUCSC.pl#L102 / ...

[Synanth]

Thank you for the resources.

Answer 2

[Maximilian Haeussler]

What is this software? Most likely, it will just ignore the "bin" column, you can put 0 into that, I bet.

As for the other columns, I think there is one column in the new format that exists also in the old format. The columsn chromStart chromEnd correspond to txStart and txEnd.

In short: use awk to just reorder the columns.

Please let us know if that works.

[Synanth]

It's transposable element quantification software. I don't think bin is necessary due to some of the authors responses to issues n github. Someone else provided a column translation. Thank you.