Conversion from base space to colorspace

[KevinLam]

Hi is there a script that can do this?

Hypothetically, I wish to convert known ESTs to colorspace so that I may assemble short reads of a transcriptome with known ESTs in a colorspace aware software.

Thanks!

[DNAjunk]

How about the corona_lite script:

encodeFasta.py myRead.fasta

output is in color space

Ex.

>myRead.fasta
CGCCTTGGCCGTACAGCAGGGGCTTACGAA

>myRead.csfasta
C33020103031311231200032031320

[aguffanti]

Why not doing the opposite, ie convert color space in sequence space usign the usual rules and after using a standard assembler? I would also ty the SAET tool which has been just posted on the SOLiD software web page on your color space reads and then convert in color space. Quick note: I have many problems in getting encodeFasta.py working correctly

[SOLiD J]

Why not doing the opposite, ie convert color space in sequence space usign the usual rules and after using a standard assembler? I would also ty the SAET tool which has been just posted on the SOLiD software web page on your color space reads and then convert in color space. Quick note: I have many problems in getting encodeFasta.py working correctly

There's a thread on this topic, but in short you get the most out of color space by aligning in color space. I believe its especially important for SNPs and catching sequencing errors.

[aguffanti]

color space to base space

Indeed working in CS does have advantages in various context, from miRNA analysis to, correctly, variation detection. However, the question here was specifically on assembly I understood.

I would definitively go, in that case, for SAET + EncodeFasta.py (used with -d option i think) + a good sequence space assembler. This is also more or less the way in which the ABI port of Velvet works.

You may also try, if you have a good reference sequence, with the Mosaik assembler from Marth lab which handles both CS and Sequence Sapce together

HTH, Regards,

Alessandro

[flobpf]

A very basic script

I couldn't find any code for doing this, so wrote a quick python script. I've validated it using a couple of test sequences in the Data Formats file from ABI, but I must warn you...use it at your own risk!

This script requires sequence in one continuous line after the name. Interleaved sequence, split across multiple lines needs to be first merged into one single line.

import sys
'''
Combinations (FROM CMS_058717.PDF):
AA:0, AC:1 ,AG:2 ,AT:3
CA:1 ,CC:0 ,CG:3 ,CT:2
GA:2 ,GC:3 ,GG:0 ,GT:1  
TA:3 ,TC:2 ,TG:1 ,TT:0
'''
print "INP1: FASTA sequence file"

#########
def base2color(basepair):
    bdict={'AA':0, 'AC':1 ,'AG':2 ,'AT':3, 'CA':1 ,'CC':0 ,'CG':3 ,'CT':2, \
           'GA':2 ,'GC':3 ,'GG':0 ,'GT':1, 'TA':3 ,'TC':2 ,'TG':1 ,'TT':0}
    return str(bdict[basepair])
#########

file1=open(sys.argv[1],'r')
out1=open(sys.argv[1]+".color",'w')
line1=file1.readline()
while line1:
    if line1.startswith('#'):
        pass
    else:
        if line1.startswith('>'):
            name=line1.strip()[1:]
            line1=file1.readline()
            seq=line1.strip()
            clist=[]
            for i in range(0,len(seq)-1):                                
                b1=seq[i].upper(); b2=seq[i+1].upper(); bp='%s%s'%(b1,b2)
                #print name,i,b1,i+1,b2,len(seq)
                color=base2color(bp)
                clist.append(color)
            out1.write('>%s\n%s\n'%(name,''.join(clist)))
    line1=file1.readline()
file1.close(); out1.close()
print "Done!"