NOVO2MAQ Bus error???

[noah]

Hello all,
I am a computer rookie, so I'm not completely sure what I am doing, but here goes.
-I am trying to look at editing events in RNA using the maq. (similar readout as SNP analysis)
-My samples are from C. elegans
-I have aligned my reads using NOVOalign.

I am trying to use NOVO2MAQ to convert my novo txt file to .map files. I have tried both the version downloadable from NOVOCRAFT ("fixed") version and version 0.7.1 of maq. Both give similar errors. The "error" being a Bus Error. I am wondering if something is wrong with the formatting of my novo.txt file or my "in.list" file. my in.list file looks like:

197-72:~ noah$ cat /Users/noah/Desktop/in.list
chrI chrI

and my novo.txt file looks like:
>8_1_1749_1647 S ACAGGTCTGTGATGCCCT IIIIIIIIIIIIIIIIII R 0 3.00871 234 >chrI 15063450 F . . .
>8_1_1749_1647 S ACAGGTCTGTGATGCCCT IIIIIIIIIIIIIIIIII R 0 3.00871 234 >chrI 15070647 F . . .
etc........

I can shorten my list of novo reads to around 105 and the bus error goes away.
Any ideas????
Thanks in advance

[noah]

I'll post this for the other noobs like myself.
After talking with Colin (NOVOCRAFT), looks like my issue came from not having the "header/command line" info in my NOVO files. I'll post what Colin sent me and the Patch for NOVO2MAQ that will circumvent the issue in case anyone else is having problems:
From Colin:
"When you use the -m option in Novoalign it adds an extra score for a possible hairpin structure to each report line. This extra field pushes the following fields across by 1. Novo2maq allows for this
but it's detection method looks at the first few lines of the file to pick up the command and see if the -m option is present.
# novoalign (2.01.06MT) - short read aligner with qualities.
# (C) 2008 NovoCraft
# Licensed to Novocraft Technologies Sdn Bhd
# novoalign -m -d /wd5/db/pgt -f 10000_75mers_Rust_prb.txt -t 90
# Interpreting input files as Solexa prb.txt.
# Index Build Version: 2.1
# Hash length: 14
# Step size: 1

So to get your file converted I had to add the command line and then it converted OK. Without the -m option the program seg faults.

Here's a patch to novo2maq source so that it detects the miRNA hairpin score on a line by line basis rather than from the '# novoalign' line.

Insert the red lines at 288,289 in novo2maq.cc. I'll also add this to next release.

for(i++;i <= NLAST; i++ )
novo2[i] = mtstr;
}
else
miRNAmode = novo[N_offset][0] == '>' ? 1 : 0;
// read flag
if (novo[n_offset][0] != '\0' || (paired && novo2[n_offset][0] != '\0')) { // there is an alignment
// set name