Can RMSIP include trivial modes for calculation?

[Yazhini arangas]

Does any analysis on normal modes in Bio3D include first 6 trivial modes for calculations?. If so how can I select only specific range of modes from nma object for RMSIP, RMSF and other trajectory analysis. It is mentioned that by default RMSIP will consider first 10 modes. Is it from modes of 7-16 or 1-10. How can I select specific range of modes for calculations? I referred subset option in RMSIP which is for only to specify number of modes.

Thanks in advance

[Xinqiu Yao]

All calculations using NMA modes have excluded the 6 trivial modes by default (and should do). Most functions provide a way to select the range of modes. For example, rmsip(..., subset=20) (use the first 20 non-trivial modes for RMSIP), fluct.nma(nma, mode.inds=c(7:16)) (use the first 10 non-trivial modes for fluctuation calculation). Please see corresponding help documents for more detail.

[Yazhini arangas]

Yes I understood that It has to remove since trivial modes will have zero frequencies and eigenvalues. I was bit skeptical about this and now I am confident and understood much clearly from your response. Another point, for comparing two different proteins' normal modes (i.e overlap, RMSIP, difference in RMSF calculation) is there any in-built normalization function available?. Or we have to individually normalize (by Z-score or divided by maximum frequencies) each protein and compare them?

Thank you.

[Lars Skjærven]

Hi Yazhss, The mode vectors are already normalized prior to the RMSIP calculation. The squared fluctuations are not normalized. see e.g. http://starship.python.net/pipermail/mmtk/2009/001573.html and other threads from MMTK.

[Barry Grant]

Q&A