Running macs with negative -G

[GoogleCodeExporter]

Hi Gary

Thanks for your help in getting macs set up on my Mac last week. I am now 
trying to simulate a chromosome under different demographic scenarios. All goes 
ok until I try to simulate an exponentially shrinking population using a 
negative -G (as you would in ms) . In this case the run either hangs, or aborts 
with errors like this:

Tree:112,pos:2.19213e-05,len:1.19761,TMRCA:0.11807,ARG:,len:1.60599,TMRCA:0.1343
24
proposed grandMRCA != top Node
proposed grandMRCA != top Node
iHistoryMax: 88
Tree:113,pos:2.19539e-05,len:1.21005,TMRCA:0.11807,ARG:,len:1.63406,TMRCA:0.1343
24
proposed grandMRCA != top Node
proposed grandMRCA != top Node
iHistoryMax: 89
Tree:114,pos:2.33874e-05,len:1.23026,TMRCA:0.11807,ARG:,len:1.64076,TMRCA:0.1343
24
proposed grandMRCA != top Node
proposed grandMRCA != top Node
proposed grandMRCA != top Node
iHistoryMax: 90
Tree:115,pos:2.47017e-05,len:1.26891,TMRCA:0.11807,ARG:,len:1.63976,TMRCA:0.1343
24
proposed grandMRCA != top Node
proposed grandMRCA != top Node
proposed grandMRCA != top Node
iHistoryMax: 91
Tree:116,pos:2.48832e-05,len:1.26891,TMRCA:0.11807,ARG:,len:1.6309,TMRCA:0.13432
4
proposed grandMRCA != top Node
iHistoryMax: 92
Assertion failed: (px != 0), function operator->, file 
/Users/samanthaoloughlin/boost_1_51_0/include/boost/smart_ptr/shared_ptr.hpp, 
line 424.
Abort trap: 6

In this case the command used was:
./macs 26 41e6 -T -t 0.008 -r 0.08 -h 1e2 -G -8.605951 -eN 0.607360 1.49  > 
outfile

Exactly the same command with a positive value for -G works fine.

Is there a fix for this?

Cheers

Sam

What steps will reproduce the problem?
1.
2.
3.

What is the expected output? What do you see instead?

What version of the product are you using? On what operating system?

Please provide any additional information below.

Original issue reported on code.google.com by DrSamm...@gmail.com on 5 Nov 2012 at 11:47

[GoogleCodeExporter]

Sam,

I'm sorry, but it doesn't seem that macs will be able to support a negative 
growth parameter because of the way the ARG is constructed (which is different 
from ms).  At least ms should work for you right?  I will at least put a 
warning and stop the program in this version.

Thanks

Gary

Original comment by gche...@gmail.com on 5 Nov 2012 at 10:07

• Changed state: Started

[GoogleCodeExporter]

Hi Gary

Thanks for letting me know. I will persist with ms; current simulation with 10 
samples has been running for 18 hours...

Sam

Original comment by DrSamm...@gmail.com on 6 Nov 2012 at 11:18

[GoogleCodeExporter]

Sorry to hear that.  I will keep trying to see if there is a workaround, but I 
can't promise anything.  Thanks.

Original comment by gche...@gmail.com on 6 Nov 2012 at 3:06

[GoogleCodeExporter]

I don't think we'll be able to support this feature for now.  Suggested using 
ms for now as workaround.

Original comment by gche...@gmail.com on 14 Nov 2012 at 6:14

• Changed state: WontFix