dear sergei
I again, then run  GARDProcessor.bf show error:

(echo data/demo.seq; echo /home/jlacosta/Download/programas/HYPHY_Source/GARD_demo.out;) | mpirun -np 2 HYPHY TemplateBatchFiles/GARDProcessor.bf

/home/jlacosta/Download/programas/HYPHY_Source/TemplateBatchFiles/Please load a nucleotide data file::
/home/jlacosta/Download/programas/HYPHY_Source/TemplateBatchFiles/Please load a GA partition analysis result file::
Loaded 1023 partitions


Sequences :4
Sites     :273
Variable  :44


f(A) = 0.458791
f(C) = 0.141941
f(G) = 0.159341
f(T) = 0.239927


c-AIC = 1180.51
Log Likelihood = -576.553622490902;
Shared Parameters:
GT=0.0702169=0.0702169
CT=0.637412=0.637412
betaP=0.856559=0.856559
CG=0.133953=0.133953
AT=0.201307=0.201307
AC=0.388101=0.388101
alpha=0.0234925=0.0234925
betaQ=2.97442=2.97442

Tree givenTree=(a:0.0326137,b:0.00689777,(c:0.0266298,og:0.0874902)Node3:0.0488684);

Mean divergence : 10.9395%


Fitting a single-tree, multiple partition model
mpirun noticed that job rank 0 with PID 5382 on node FIERA exited on signal 11 (Segmentation fault).
1 additional process aborted (not shown)


??????

sorry!!!

Dear Luis,

I think you may be using an old vertion of GARDProcessor.bf
The line

[code]
Loaded 1023 partitions
[/code]

tells me that the code is not reading the number of partitions correctly.
Try using GARDProcessor.bf from the Standard Analyses menu (under Recombination). I attach a folder containing a simple example.

Cheers,
Sergei

---

dear sergei

whole procedure was::
$  ( echo ; echo data/demo.seq; echo ; echo 010010; echo 1; echo \n;) |  mpirun -np 8 HYPHYMPI TemplateBatchFiles/GARD.bf

show it:

GA has considered        355/       13244 (1230 over all runs) unique models
Total run time           0 hrs 1 mins 25 seconds
Throughput               14.47 models/second
Allocated time remaining 999 hrs 58 mins 35 seconds (approx. 5.20929e+07 more models.)

GENERATION 50 with 3 breakpoints (~96% converged)
Breakpoints    c-AIC  Delta c-AIC [BP      1] [BP      2] [BP      3]
         0  1177.93
         1  1170.30        7.629        114
         2  1167.12        3.174        114         206
         3  1167.12        0.000        114         206
GA has considered        362/       13244 (1237 over all runs) unique models
Total run time           0 hrs 1 mins 25 seconds
Throughput               14.55 models/second
Allocated time remaining 999 hrs 58 mins 35 seconds (approx. 5.23894e+07 more models.)

GENERATION 51 with 3 breakpoints (~98% converged)
Breakpoints    c-AIC  Delta c-AIC [BP      1] [BP      2] [BP      3]
         0  1177.93
         1  1170.30        7.629        114
         2  1167.12        3.174        114         206
         3  1167.12        0.000        114         206
GA has considered        369/       13244 (1244 over all runs) unique models
Total run time           0 hrs 1 mins 26 seconds
Throughput               14.47 models/second
Allocated time remaining 999 hrs 58 mins 34 seconds (approx. 5.20732e+07 more models.)

GENERATION 52 with 3 breakpoints (~100% converged)
Breakpoints    c-AIC  Delta c-AIC [BP      1] [BP      2] [BP      3]
         0  1177.93
         1  1170.30        7.629        114
         2  1167.12        3.174        114         206
         3  1167.12        0.000        114         206
GA has considered        376/       13244 (1251 over all runs) unique models
Total run time           0 hrs 1 mins 26 seconds
Throughput               14.55 models/second
Allocated time remaining 999 hrs 58 mins 34 seconds (approx. 5.23662e+07 more models.)
Performing the final optimization...
I generate these files:
messages.log.mpinode0
messages.log.mpinode1
messages.log.mpinode2
messages.log.mpinode3
messages.log.mpinode4
messages.log.mpinode5
messages.log.mpinode6
messages.log.mpinode7
where is gard.splits ???

I take the last file that generates GARD.bf

(echo data/demo.seq; echo /home/jlacosta/Download/programas/HYPHY_Source/messages.log.mpinode7;) | ./HYPHYMP TemplateBatchFiles/RecombinationProcessor.bf

show me:
TemplateBatchFiles/RecombinationProcessor.bf

/home/jlacosta/Download/programas/HYPHY_Source/TemplateBatchFiles/Please load a nucleotide data file::
/home/jlacosta/Download/programas/HYPHY_Source/TemplateBatchFiles/Please load a GA partition analysis result file::
Loaded 2970.5 partitions

Sequences :4
Sites     :273
Variable  :44


f(A) = 0.458791
f(C) = 0.141941
f(G) = 0.159341
f(T) = 0.239927


c-AIC = 1180.51
Log Likelihood = -576.553622490902;
Shared Parameters:
GT=0.0702169=0.0702169
CT=0.637412=0.637412
betaP=0.856559=0.856559
CG=0.133953=0.133953
AT=0.201307=0.201307
AC=0.388101=0.388101
alpha=0.0234925=0.0234925
betaQ=2.97442=2.97442

Tree givenTree=(a:0.0326137,b:0.00689777,(c:0.0266298,og:0.0874902)Node3:0.0488684);

Mean divergence : 10.9395%


Fitting a single-tree, multiple partition model

my question is how to generate file GARD.split???

Sorry for the inconvenience and thank you in advance.

Dear Luis,

GARD.bf prompts you (the last echo statement in your example) for the location of the results file. That file will be written by GARD along with several other files in the same path. On of them will end with _splits - that's the one you want to feed to the processor. E.g. if you request that the results be written to /home/user/results/main then the splits file will be in /home/user/results/main_splits

Simply edit your echo statements to include a non-trivial path and all should go well.

Cheers,
Sergei

PS message.log files only contain diagnostic messages about the run

dear  sergei!!
everything is perfect! thank you very much!!!