Spending time learning the most basic commands from information gleaned on FAQs and other help pages is not gratifying. Even the vocabulary is abstruse – one “compiles” rather than “installs” and where the heck was that zlib.h, in the end? I’ve apt-got lots of stuff.
But then again, decalcifying tissues for future embedding, tracking down lent-out antibodies in the building and a number of other wet-science-related tasks today were not much more exciting.
After my third install of Ubuntu 9.04 on my home computer, I think I have it. I might finally be able to run Maq, and possibly Bowtie, should I so wish. Thanks in part to Richard for getting me over the hump by believing I might have the ability to deal with such things.
These programs will allow me not to rely only on the goodwill and free time of collaborators, but actually map many millions of short sequences to a reference genome sequence. The human one. While it is not the largest genome out there, it’s plenty enough to overwhelm the ability of my laptop’s Windows to deal with the 2.9 Gb of text in the human reference genome available here. So I had to learn how to use a Linux-based operating system that somehow does something nifty with 64 bits. It has been explained to me but it just will not stay in my head.
I had been meaning to take the plunge for a long while, but who has multiple full days to learn a new way of doing what one managed to do earlier? It entailed a situation in which it was not possible to manage, to push me over the edge. I had to find out how to stick lots of text (from each chromosome) together – that would be by a series of commands along the lines of:
$ cat chr*.fa >> ref.fasta
Lots of people in my acquaintance could have told me how to do that. But I found out myself, which is much more fun when it finally works. Here it is, already midnight, and I have also found out how to cut and paste in a terminal window, and a few other goodies. Meanwhile, Stephen is barbecuing on a Friday night, and I am a wee bit jealous.
Finally, I also found out that I downloaded the wrong version of the human genome, and there is a new reference that has been available the last few months. So, instead of seeing whether or not my Maq program is really capable of running, I’ll have to go to sleep and leave my computer to acquire a new human genome.
I wonder if the two versions might undergo some sort of homologous recombination during bedtime?
It burns.
I left a witty and incisive comment here, and it’s disappeared.
Damn.
It was probably the wrong version.
(slaps forehead) Yes, I bet I see where that comment would have come in. Sigh. Bit late to edit now, eh?
CVS. You know you want to.
What?
What? Have we been upgraded? Damn.
Ah well, I guess you understand this now:
Indeed. On one of the help forums, someone had paraphrased that great classic in their signature, except in the typical geek sexist manner, it went along the lines of:
MAN: Make me a sandwich.
WOMAN: What? Make it yourself.
MAN: Sudo make me a sandwich.
WOMAN: Okay.
Quite a bit less funny.
@Richard – I thought you were referring to my mentioning hump, a fellow Richard, and homologous recombination all in the same post.
Might have been. It was another country, and besides, the wench is dead.
Out, out, damned spot!
No, wait…