How to interpret DEGs of DESeq2 in comparing more than two groups?

Question

[takoyaki]

Hi !

I want to ask how to interpret DEGs in multiple group comparison by DESeq2.
Now I have three groups ( A vs B vs C ) and run likelihood ratio test like below.

dds = DESeqDataSetFromMatrix(countData = count, colData = group, design = ~ condition)  
dds = DESeq(dds, test="LRT", reduced=~1)

Next I checked each result groups like this.

resultsNames(dds)  
[1] "Intercept"        "condition_B_vs_A" "condition_C_vs_A"  

res = results(dds, format = "DataFrame", name ="Intercept" )
res = results(dds, format = "DataFrame", name ="condition_B_vs_A" )
res = results(dds, format = "DataFrame", name ="condition_C_vs_A" )

Below are my questions.

1. I think my code run likelihood ratio test between full model and reduced model.
   In this case, if p-value is lower than threshold, it only means that this gene expressed differentially in any of groups, but we don't know that this gene shows whether high or low expression in which group. Do you think my understanding is correct ?

2. We get fold changes by selecting comparison group from results function. So, we can get Up DEGs or Down DEGs by comparing specific two groups though we calculated p-value by comparing all groups. Is that right ?

3. Do you know how to get Up / Down DEGs that was higher or lower in any of group not specifying comparison two groups in DESeq2 ?

p.s. I have read this post and this is really helpful, but I wanted to know more. https://www.researchgate.net/post/How_to_interpret_results_of_DESeq2_with_more_than_two_experimental_groups

thanks !