I really like the approach taken in the SCDE package for single cell differential expression. In the examples differential expression analysis is performed between two cell types but would it be appropriate to use this package for differential expression in other scenarios? For example, all cell types at two time points? Or just cells of the same type at two time points? 

i.e 

Cell type A, B, C  at T1  vs  A, B, C at T2

or

A at T1  vs  A at T2

Also, why is it that we get different results when running 

scde.expression.difference and scde.test.gene.expression.difference?

Here is an excerpt from their example:

##           lb     mle       ub       ce        Z       cZ

## Tdh 5.807793 8.075220 10.302866 5.807793 7.159589 5.989598

## Tdh 5.728235 8.03544 10.30287 5.728235 7.151425 7.151425

And with my data:

##           lb     mle       ub       ce        Z       cZ

EEF1A1          -1.304039   -1.188977   -1.112269   -1.112269   -7.160809   -7.160809

EEF1A1          -0.53693685   -0.421879   -0.3451737   -0.34517369   -7.160847   -6.123877