Hello,
I computed semantic similarity score between gene ontology terms, which ranges from 0 to 1. I want to choose cut-off in these 0 to 1 value. By which I can choose similar and dissimilar GO terms based on this cut-off value. Can somebody suggest how to determine significant cut-off.
Thanks,
Sainitin
It is depending on the method you use for example the ERIC PCR you would choose %85 ,but for techniques like PFGE it would be %95 because it is more consistent ,but you always look at banding pattern ,some bands have higher intensity and not always picked up by software
Hi Sainitin,
Have a look at the PCIT r package. It's designed to eliminate "spurious" correlations. If you treat semantic similarities between terms as "correlations", you might have some luck in determining the best cut-off value.
Article PCIT: An R package for weighted gene co-expression networks ...
- Badges
- Science topic
- Similar topics
- Biological Science
- Molecular Biology
- Molecular Biological Techniques