Hello everyone
I performed weighted unifrac on 16S rRNA data. I have very strange PCoA plot this time where my Axis 1 is 93% and Axis is 5% and 3 is less than 1%. All samples are clustered together at the base of the one axis. What does it indicates if my Axis 1 is more than 90%? I attached file here
Dear Kévin Vervier it a strange situation. I am using emperor tool for visualization of this graph. Axis 1 is so long that it only fits into the screen. If I make 3d graph, I can rotate it but if I make 2d plot, I can only see a long horizontal line (Axis 1) on the screen.
I guess it simply means 91% variability in your data is represented by Axis 1. In other words your data does not contain much variability.
Dear Prashant Singh I agree. thats what clustering pattern and PERMNOVA also showing. But I am guessing that there may be more information than this. Like in past I remember that a greater value of the sum of all three axis is considered to present more diversity.
I am not sure if adding three axes (given only one axis has >90%) is confirmation of variability in data. If the data scattering is like 20%, 30% , 40% in each axis, then you could say there is definite variability to look at.
Try changing or adding the new parameters in mapping file.
Hope it helps.
- Badges
- Science topic
- Similar topics
- Engineering
- Petroleum Engineering
- Oil