forked from MikeAxtell/CleaveLand4
-
Notifications
You must be signed in to change notification settings - Fork 0
/
README-Threaded
42 lines (28 loc) · 1.69 KB
/
README-Threaded
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Readme for threaded version.
==============================
Created by: Bruno Costa
ITQB 2016
What this is:
This is a wrapper for Cleaveland 4
The main objective of this process is to speed up the process of target prediction with cleaveland.
The implementation is perfectly scalable as long as you have the necessary RAM to allocated the transcriptome "N" times into memory.
The paralazation is done by. Splitting the sRNA fasta file into smaller portions. Say 10 sequences per thread. Cleaveland is run "N" times in separate subdirectories.
Once a process is finished the wrapper grabs the next 10 sequences in the original sRNA fasta file.
The number of sequences to be processed by each thread is customizable.
Once the whole fasta file has been parsed the program builds a unique output
Dependencies
============
Must have the extract_x_seqs.sh and fragment_abundance.py in your path. It comes with this release (See Fasta_extractor)
Set up the location of Cleaveland4.pl (should be solved soon to relegate to path as well)
So far all I can remember.
All the others for Cleaveland4. Cleaveland4 must be able to run on it's one for this to run.
Valid modes:
Only mode 1 and mode 2 are setup so far.
Mode 1 run a few sRNA sequences inorder to buid a denisty file (Sorry was to lazy to check how the density file is produced.) Afterwards passed the degradome density file to run in mode 2 in a parralel way.
-p Pvalue Default 1
-r MFEratio default 0.65
-g gstar might not be implemented because it is for mode 3/4
-q quiet mode will be implemented shortly
-c Category default 4
-a in thought. I'll think about it.
Comments or suggestions open issue on github or send and email to: [email protected]