bowtie error "Error reading _offs[] array: "

[Nicolas902]

Dear all,

I am struggling with a bowtie error which occurs in the ebwt.h, here is the code:

else {
						// Do it all in one read
						MM_READ_RET r = MM_READ(_in2, (void*)this->_offs, offsLen << 2);
						if(r != (MM_READ_RET)(offsLen << 2)) {
							cerr << "Error reading _offs[] array: " << r << ", " << (offsLen << 2) << endl;
							throw 1;
						}
					}

I have built an index using a regular fasta file of hg18 downloaded form ucsc and a splice.fa file generated using ERANGE. bowtie-build is running without any error, as well as "getsplicefa.py" which generates the splice file.

Any idea what this error means?

Thank you.

[anecsulea]

Hi,

I've been having a similar error. I'm building a bowtie index for a genome sequence, bowtie-build finishes without error, but when trying to align reads with bowtie I get this error:

"Error reading ebwt array: returned 41750080, length was 168445184
Your index files may be corrupt; please try re-building or re-downloading. "

Have you found a solution to the problem ? All suggestions would be appreciated.

Best,

Anamaria

[Nicolas902]

If I remember well, this was a question of directory. I moved the chromosomes sequences I used to the "bowtie" directory.

I hope this was the solution, tell me if it makes it for you.

[anecsulea]

No, this can't be it... For me, this error occurs while running TopHat, and in the TopHat run the fasta file and the bowtie index files are in the same directory.

[Ms@New]

Hi, I get the same error, and I'm using hg18 index from bowtie website.

Did you figure out what was the problem?

[anecsulea]

I don't know what the problem was, but I think I did find a solution: see

http://seqanswers.com/forums/showthread.php?t=5318

Let me know if you need more help on this.

Hi, I get the same error, and I'm using hg18 index from bowtie website.

Did you figure out what was the problem?