Segmentation Fault in fda_view_stress

[aminsagar]

Dear fda developers,
I am trying to use gromacs-fda to calculate punctual stress during the course of a steered MD simulation.
My input file looks like this

onepair = summed
atombased = no
residuebased = punctual_stress
type = all

gmx_fda mdrun completes without any error and the output .psr files are written.
However, when I try to generate PDB files using
gmx_fda fda_view_stress -i output1.psr -s ../structure.tpr -f ../traj.xtc -n ../index.ndx -o stress1.pdb -frame 1 -convert yes
I get segmentation fault.
When it asks for residue model points, I have tried Protein and C-alpha but none works.
The final output looks like this

stress filename = output1.psr
result filename = stress1.pdb
frameType = single
frameValue = 1
stressType = punctual
particleType = residue
convert = 1
Reading file ../WT-100pN-500ns-1.tpr, VERSION 2020.4 (single precision)
Reading file ../WT-100pN-500ns-1.tpr, VERSION 2020.4 (single precision)

Back Off! I just backed up stress1.pdb to ./#stress1.pdb.3#
Segmentation fault (core dumped)

I would really appreciate any help/suggestions.

Best,
Amin.
Center de Biochimie Structurale,
Montpellier,
France.

[BerndDoser]

Hi Amin,

thanks for your report and your patience. I am now back from vacation and will take a look to your issue. Could you please send me your input files and the FDA version number you are using, so that I am able to reproduce your issue.

Best regards

[aminsagar]

Hello Bernd. Thanks for your response.
I am attaching the files for reproducing the error.
I am using fda 2020.
fda-test.zip
Best,

[BerndDoser]

Thanks for the files. The pdb output format is only working if the input structure file -s is also in the pdb-format. Otherwise the pdbinfo is not available in gromacs. I will try to generate the missing pdbinfo. If this is not possible, I will restrict the input format to pdb. However, I found a second curious issue in your input data, which I would like to figure out. Your system has 102 alpha-C atoms, whereas the stress.psr file contains 103 entries with the first entry is always zero. Could you please also send me the input files to reproduce the generation of the psr-file. Thank you very much.

[aminsagar]

Thanks Bernd for the help and sorry for replying late. I confirm that pdb file is generated correctly on using the pdb file.
I noticed that there was an extra column in the psr file but ignored it. I might be completely wrong here but it seems to me that this happens only when I have "residuebased = punctual_stress" in my input file. I regenerated the psr file with a pfr file generated with "residuebased = pairwise_forces_scalar" using fda_view_stress and fda_get_stress and I didn't get the extra column.
I have attached my initial psr file (using residuebased = punctual_stress) which I used to get the stress.psr file (using fda_get_stress).
input-and-intiialpsr.zip