Random Forest implemented in python.
Wiki explanation gives the basic.
A collection of Machine Learning methods

How to install package “sklearn.ensemble”

I am learning Python using kaggle titanic project, the first error I got was, how can I install a package.

Here, is my answer under MacOX

It turns out cuffdiff gives much different results from HT-Seq/DEseq. So, I am trying to use HT-Seq/DEseq for Thuy-Ai’s project

Here is the documentation for using HT-Seq

There are 14 columns in cuffdiff output

awk -F"\t" '{print $1"\t"$8"\t"$9"\t"$10}' gene_exp.diff | head 
test_id value_1 value_2 log2(fold_change)
A2M-AS1|144571|locus1of1|1      0.0190991       0.0447137       1.22722
A2MP1|3|locus1of1|1     0       0       0
A2M|2|locus1of1|1       0.0261932       0.0171011       -0.615099
A3GALT2|127550|locus1of1|1      0.0287783       0       -1.79769e+308
A4GNT|51146|locus1of1|1 0       0       0
AA06|100506677|locus1of1|1      0       0       0
AACSP1|729522|locus1of1|1       0       0.00822783      1.79769e+308
AADACL2-AS1|101928142|locus1of1|2       0       0       0
AADACL2|344752|locus1of1|1      0       0       0
...

the log2(fold_change) was computed by log2(value_2/value_1)

where “value_1” comes from the first “bam file” and “value_2” comes from the second “bam file” as the following command

cuffdiff  -o /somePath/ /somePath/transcripts.gtf value_1_firstBam.bam value_2_secondBam.bam

Here is an old link

If the pairs are overlapped, they could be merged

It seems that the alingers do NOT care about the overlap and align them anyway. For tophat, it even allows negative insert input. It might not be relevant too much with RNAseq, where the read will be counted once toward the quantification. But, it will matter if one wants to detect peaks esp. display peaks on the broswer.

Step 1, I need to create a repository on my github central place

	
Create a new repository called 'rShiny' at github
	
git clone https://github.com/2waybene/rShiny rShiny under "myGit"
	
I have an account with basic feature at shinyapps
	
This one was for Dr. Cui, his project on MotionDetection

Step 2 RShiny tutorial

	
Here is the main tutorial page from RStudio
	
And, there are many examples in this gallery.

Some special RShiny note

	
Install shinyIncubator needs devtools

To start shinyapp on the command line, use the following command

R --vanilla -e "shiny::runApp('.', port=8888, host='10.91.128.1')"
R --vanilla -e "shiny::runApp('.', port=8888, host='localhost')"

Then go to a browser to load the application. For some reason, it fails to work with the follow error message

Arraystar provides services