Luca Beltrame
64b24842b8
All checks were successful
continuous-integration/drone/push Build is passing
22 lines
1.4 KiB
Markdown
22 lines
1.4 KiB
Markdown
---
|
||
author: einar
|
||
categories:
|
||
- Science
|
||
comments: true
|
||
date: "2007-07-12T19:20:55Z"
|
||
header:
|
||
image_fullwidth: banner_other.jpg
|
||
slug: full-speed-ahead
|
||
title: Full speed ahead
|
||
omit_header_text: true
|
||
disable_share: true
|
||
wordpress_id: 271
|
||
---
|
||
|
||
After giving in the summary of my thesis today, I also started writing the actual thing. Of course this is not something I will finish in a few days. It's a long journey that will go on until I deliver it.
|
||
|
||
Aside that, I've been having some problems with R, a language which I really don't like and I hope to use as less as possible. I was obtaining a list of differentially expressed genes (DEGs) out of some data files with [SAM](http://www-stat.stanford.edu/~tibs/SAM/), and of course I had to supply a matrix with expression values for every gene.
|
||
|
||
What I _didn't_ understand is why on Earth the package [siggenes](http://bioconductor.org/packages/2.0/bioc/html/siggenes.html) understands that, but when outputting the result (command sam2excel) it's giving the **row** of the file where the gene is, instead of the actual gene. If I supplied the IDs, whatever they are, they should be added as well! But you can output those only with "list.siggenes" which of course doesn't add the d-score I need...
|
||
|
||
In general, I find R slow, clumsy and counter-intuitive. There are much better languages out there (and with better interactive shells), I can't really understand why it is so popular among people who do bioinformatics...
|