What are acceptable results in rsem-plot-model?

[Dekel Wainer]

Hi all!

I performed rsem-calculate-expression on 4 samples and received what seems to me low-quality alignment. 

Where do you usually place the limit and say that a sample is too low quality to continue analysis? Is there something I can change in the run?

Here are 2/4 samples' alignment statistics:
1. Unique: 26%; Multi: 41%; Unalignable: 33%

2. Unique: 20%; Multi: 31%; Unalignable: 49%

Thanks a lot!