Comments on: Dreaming about bio-spreadsheet

By: Pawel Szczesny

Pawel Szczesny — Mon, 22 Jun 2009 11:28:51 +0000

Thank you all for the comments.

Marcin, it’s all about presentation to the non-programming colleagues and looking for possibility to avoid Excel files .

Burke, thanks a lot, I wasn’t aware of ISAcreator. Need to have a look.

Tyler, duh!, DableDB I’ve used it so long ago I forgot about its existence.

Bogdan, I have a similar workflow (but use Perl one-liners instead of gawk) however as I pointed to Marcin, it’s the “presentation” part that dictates other requirements . I’ve corrected your gawk snippet.

By: Bogdan

Bogdan — Fri, 05 Jun 2009 08:57:21 +0000

oops, missed a closing ‘ after right curly brace above

By: Bogdan

Bogdan — Fri, 05 Jun 2009 08:56:38 +0000

I’m mostly stuck with Python doing this kind of things. Each time I just copy over the older script, change the number of items parsed from a single line of data file, and then change print statements

That’s fairly easy, but I guess this could be extended to a more-or-less universal script, counting columns on its own, and accepting string parameters with $1, $2 … embedded to denote column contents.

Also, there’s awk (gawk), which I guess could do the job best, but my knowledge of gawk is limited to simple column extraction, like
gawk '{print $2"\t"$5}' datafile.tsv

By: Tyler

Tyler — Sat, 23 May 2009 20:30:30 +0000

It seems like you could probably throw something together with dabbleDB that would fit the bill…

By: Burke Squires

Burke Squires — Wed, 20 May 2009 16:18:04 +0000

I am not sure is this fits the bill…but some folks at the EBI have created a java-based platform for data storage and annotation. It is called ISAcreator and can be found here:

http://isatab.sourceforge.net/isacreator.html

Burke

By: Marcin Cieslik

Marcin Cieslik — Tue, 19 May 2009 14:10:18 +0000

So data is in rows and columns are “transformations”. Why do you think this is superior to a flow-chart with vector(i.e. rows) input?