In pursuing comparative genomics approaches for functional inference of protein function, the initial selection of related proteins separated by great evolutionary distances can be a challenge.
See Figure 3. FlowerPower is designed for the first step in structural phylogenomic inference of protein function: selecting a dataset upon which functional inference will be based.
Then, Gibbs sampling is employed for functional inference of the proteins with unknown function ("unannotated proteins").
Therefore, the indirect functional inference should correctly reflect the actual function.
Transferring annotation of proteins at a domain level allows more accurate functional inference [ 21] and is useful for predicting the function of multi-domain proteins [ 22] and novel domain combinations that possibly give rise to new protein functions [ 23].
Finally, functional inference of these core stress responsive proteins using both sequence similarity and non-similarity approaches was conducted.
We described here a global detection and functional inference of hypothetical proteins involved in stress response in Synechocystis sp. PCC 6803.
Consequently, reducing complexity and redundancy in sequence databases may negatively affect efficient functional inference.
The sequence similarity feature included two scores: pair-wise similarity and cross-species functional inference.
Therefore, we applied a global network diffusion method to integrate all ETA matches into the functional inference.
This proves that using a simple sequence similarity approach like BLAST for functional inference may not be so immediate nor trivial even at high sequence identity [37].
