Installation tests about emdna

[rty10]

Idea list of tests:

1. config1, IdealDNA, --free-collection
2. config2, IdealDNA, --hold-last-bp
3. config2, Olson1998, --hold-last-origin
4. config3, IdealDNA, --frozen-step=10:20, --hold-last-bp
5. config3, test.ff, --free-collection
6. config3, test.ff, --hold-last-bp

Each config should be each of the three types of initial files (reference frame, parameter, and bp-list)
No config size beyond 150-bp: consider 45-bp, 70-bp, 110-bp

Question: Should we include in the testing (i) making a force field, (ii) parsing from input types, and (iii) getting a topology file?

[nicocvn]

1. making a force field is "weakly" covered by the unit tests so that would be great.
2. Yes totally.
3. Does that mean running the topology tool that computes Lk, Wr and Tw? If that is having at least one case would be enough I think.

[rty10]

I'm glad we are in agreement on the first two.
For the third, yes, I'm referring to that tool. I've always been curious to know if there's a way, in the future, to run this in the emDNA main command line and then include Lk, Wr, Tw in the log file.

[nicocvn]

This is most likely possible. The topology tool is just wrapper around various classes in the DNASim library I believe.

Regarding test materials I might have data ... I will check over the weekend.

[rty10]

Curious to know if we can incorporate in the testing your use of the ramping function that you discussed in the text. I personally would love to learn more about this process.

[rty10]

I have four parameter files of lengths 50, 75, 141, and 150bp with three set up as circles. Does emDNA require a specific type of formatting of parameter files, such as specific column widths? I ask because I believe Pamela Perez had mentioned this and it should be addressed if a formatting condition is needed. I did run an emDNA optimization with a .par file and got a Segmentation fault.

[rty10]

Current use of emDNA-cli build on WLS2, tested the following using --DNA-seqdep-model=IdealDNA --hold-last-bp

Test set 1: "standards"

• test_50bp.par, _75bp.par, _141bp.par, and _150bp.par with --x3DNA-bp-step-params-input works. This is first time since I've started using emDNA that optimizing with a starting .par file worked.

Test set 2: "circles"

• test001.dat and test002.dat where the first frame =/= last frame with --x3DNA-bp-input works.
• test011.dat and test012.dat where the first frame = last frame (for circular forms) with --x3DNA-bp-input failed. Error:
  terminate called after throwing an instance of 'std::logic_error' what(): basic_string::_M_construct null not valid Aborted
  -test001.par and test002.par (same as test001.dat and test002.dat, but in par form) work
• test011.par (circular form of test001) worked
• test012.par (circular form of test002) did something odd- it has 151 base pairs and 150 steps. But the output says that it has 150 base pairs and 149 steps, same as test002.par. There's an error in here and do not know if it's user or code.

[rty10]

Email sent to team with 4 tests to be incorporated
Test 1- test emDNA-parser and -cli to generate a reference frame and bp list file, and then optimize all three input types with --free-collection, --hold-last-origin, and --hold-last-bp
Test 2- test emDNA-parser, -cli, and -topology to generate 3 circular parameter files from a bp-list file, each with specific sequence, optimize with IdealDNA and Olson1998 forcefields using --hold-last-bp, and then collect topology data.
Test 3- test emDNA-parser, -ff-packager, and -clit to make a reference frame file from a parameter file, make an external test forcefield, optimize with --free-collection on the test.ff, and then optimize with IdealDNA and test.ff with --hold-last-bp
Tets 4- test emDNA-cli and -topology with the --frozen-steps --hold-last-bp options only using IdealDNA. First optimize parameter file without anything frozen. Then optimize with 3 regions frozen, use opt output file as new input and optimize with 2 regions, repeat for 1 region, and repeat with nothing frozen. Finally, collect topology data of all outputs.