PC1 and PC2 gives almost identical trajectories while the vector representation (porcupine plot) gives orthogonal PC maps

[ashan dayananda]

Hello,

I am using Bio3d to visualize the PC motions of loops in a protein. When I run the PC analysis, (mktrj.pca) PC1 and PC2 give almost identical animations (slight change in the vector length). But in the pymol output, I see two orthogonal PC vector plots (which is expected). I am quite not sure why the trajectories are not showing as orthogonal. I also checked the dot product of the first two eigenvectors and it came out zero as expected. Attached are my script and outputs.

Looking forward to your comments and suggestions.Thank you

[Xinqiu Yao]

Hi,

‌

Can you create a short example data for us to reproduce your problems? Thanks.

[ashan dayananda]

  <div class="preview-container wiki-content"><!-- loaded via ajax --></div>
  <div class="mask"></div>
</div>

</div> </form>

[ashan dayananda]

Hello,

Thanks for the reply. Sure. Please find the attached files.

[Xinqiu Yao]

Hi,

‌

I’ve tried with your example. The movements of PC1 and PC2 didn’t look identical. Make sure you are using the latest version of Bio3D (2.4-1). Let me know if you still see problems after updating bio3d (if applicable).

‌

[ashan dayananda]

Hi,

Thanks. I updated Bio3D. It worked. I was using V 2.4-0 before.

[ashan dayananda]

Hi,

Additional Question,

Is it possible to add a color bar (to show the time dependence) to the conformer plots obtained from PCA?
If possible can you suggest the code?

Thanks

[Xinqiu Yao]

Not a bug