I've had better luck with 2.3b3 converting the input data files to the
unique molphy format first, then not using the -I option. They have a perl
script in the distribution that will convert from phylip to molphy.
Even so, I've only been able to get 2.3b3 to work on UNIX machines, not on
Win32, where I get similar memory errors under any condition.
Russell
--
Russell L. Malmberg
russell at dogwood.botany.uga.eduhttp://www.botany.uga.edu/malmbergleonard at datura.BCH.UMontreal.CA (Daniel Leonard) wrote in
<Pine.GSO.4.21.0009061115300.14518-100000 at datura.BCH.UMontreal.CA>:
>I have been testing protml in molphy-2.3b3.
>>I found out that the protml version in 2.3 seems broken. By running
>exactly the same test on both versions (2.2 and 2.3b3), 2.3 segfaults.
>>By examining the core dump with gdb:
>(gdb) backtrace
>#0 0x31394 in pstree ()
>#1 0x36584 in pml ()
>#2 0x35518 in main ()
>>we can see that something happens in pstree(). With 2.2, the output seems
>OK.
>>The test:
>>In gde, I selected some data then feed them to the following pipeline
>(unlerlying a menuitem):
>>itemmethod:(mkdir ./Protml.$$; cd ./Protml.$$; readseq -a -f12 ../in1 |
>sed 's/\~/-/g' > infile; protml -e -I infile > usertrees; protml -I infile
>usertrees > outfile; /bin/rm -f ../in1; textedit outfile)&
>>protml dies in the second call "protml -I infile usertrees > outfile"
>>I have not been able to find any contact adresses of the authors.
>>Has it happen to anyone ?
>>----------
>Daniel Léonard
>>OGMP Informatics Division E-Mail: leonard at bch.umontreal.ca>Département de Biochimie Tel : (514) 343-6111 ext 5149
>Université de Montréal Fax : (514) 343-2210
>Montréal, Quebec Office: Pavillon Principal G-312
>Canada H3C 3J7 WWW :
>