I have a question regaridng the conditon 

my condition is like below 

condition <- factor(c(rep("CondA", 5),rep("ConB", 5)))
> condition
 [1] CondA CondA CondA CondA CondA ConB  ConB  ConB  ConB  ConB 
Levels: ConB CondA​

what can I understand from log2fold change. does it mean that a gene from condA is higher when it is positive and or lower when it is negative  in regards to conB?

I did vignette("DESeq2") then I looked at More information on results columns which says 

log2 fold change (MLE): condition treated vs untreated" but it does not explian anything more . This is the explanation given here 

For a particular gene, a log2 fold change of -1 for condition treated vs untreated means that the treatment induces a multiplicative change in observed gene expression level of 2−1=0.5compared to the untreated condition. If the variable of interest is continuous-valued, then the reported log2 fold change is per unit of change of that variable.​

what if we use Count values?I know that it does  log2 scale but does it add +1 to it or just log2? do you safegaurd againts outliers ? do you get averge or do you get median ? 

How was it calculated ? Treated-Untreated? or Treated/untreated ? can you please comment on this ?

After reading their manuscript and manual of the package, I found that it is somehow calculated based on correlation coeficient of the GLM regression which is still a bit fuzzy to me. This is important because of the fact that I see a data where the fold change Treated/Control shows upregulation while the log2 fold change coming from Dseq2 gives down regulation and I have no idea why ? so would it be possible for someone to explian how they calculate this log2 fold change?

browseVignettes("DESeq2")