Filter frayed ends or broken H bonds

[RAHULRAH]

Hello,

I am trying to process some X-ray protein-DNA crystal data. I am just focussing on the double-stranded DNA structure. But there are some snapshots/structure in which the H-bond doesn't seem to be okay i.e. say distance of acceptor and donor atoms in H-bond is more than 5 angstrom.

Are there any justified criteria for how to filter such bad snapshots?

Thanks in advance
Rahul

[xiangjun]

Hi Rahul,

Please use concrete examples to illustrate unambiguously what you want to achieve.

Thanks.

Xiang-Jun

[RAHULRAH]

Hi Xiang,

Thanks for the reply. Sorry for not making it clear.

In simple words, how would I classify whether any H-bond in DNA is broken or how to identify base-pairs in DNA structure?

I want to use a Gaussian model similar to the one of Prof. Olson's coarse-grained model but also including intra-base parameters i.e. shear, stretch, stagger and Buckle, Propellor, opening. As you expect that some ends might be frayed due to which some of the helical parameters given by 3DNA are very weird. For example, very high opening or stretch which indicates H-bond is broken and I can see that in VMD as well.

Now the question is what shall be the criteria of such broken H-bonds? Is there any standard protocol in 3DNA? I have seen some articles using 4 angstroms as a cutoff in 3DNA, for example in, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2879501/ and https://doi.org/10.1093/nar/gky599

Thanks
Rahul

[xiangjun]

See FAQ entry: "How to fix missing (superfluous) base pairs identified by find_pair?" (http://forum.x3dna.org/faqs/how-to-fix-missing-(superfluous)-base-pairs-identified-by-find_pair/)

Again, if you provide a concrete example, we could discuss the issue in more detail.

Best regards,

Xiang-Jun