I have found that 1:3 always works. I am assuming that these are ampicillin plates. Carbenicillin plates do not grow satellite colonies because this drug is not degraded by beta lactamases. You may be incubating your plates too long. Positive colonies grow out then bacteria make beta lactamase degrading the ampicillin allowing the non-transformed bacteria to form satellite colonies. If these bacteria are accidentally picked with your positive clones they should not grow in the broth with antibiotic. You should not have to incubate plates or broth cultures more than 16hrs unless you are working with a difficult plasmid and then it would be better to drop the incubation temperature and grow your bacteria at a slower rate. Are you cloning with one restriction site or two? Do you have colonies on your vector only negative control plate? You could have vector self ligation. You could treat your vector with phosphatase prior to ligation. Are you gel purifying your digested vector? If your vector is fully digested , treated with phosphatase and gel purified then you should not get any colonies on your vector only plate. If you have too little insert than you usually end up with fewer colonies. There is never really a reason to add too much insert for sticky end ligation but could help for blunt end ligation. There is nothing unusual about your vector or insert sizes so it should work. Not sure if they are PCR products or digested/purified from a another plasmid. Either way, make sure they are clean products from gel purification and the concentration is accurate for both vector and insert.
Yes, there're satellite colonies too on the vector only negative control plate, similar to the plates with dna inserts in which they are tiny, and found mostly around the edges of the plate. The non-transformed competent cells plate, however, is clear and no colonies at all.
It seems like it could be the vector religated. But when I picked a few potential colonies from the center, and grow on broth (with antibiotic) and later extract the plasmid, there's either too little dna or no plasmid dna at all (I confirmed with gel electrophoresis). So it doesn't seem like the vector religate or else there should be shown on the gel too. It's more likely they're satellite colonies that I picked.
Anyway, I'll try to treat my cut vector with phosphatase just to be sure.
On hindsight, I think my mistake is I grow the bacteria for way too long (~20-22 hours) as the antibiotic (Erythromycin) I used is bacteriostatic instead of bacteriocidal.
I've been advised to lower the ratio as too much dna can hinder the ligation proses. But in my case. the total dna I put in the 20ul ligation mixture is not even more than 100ng, so am wondering if it's necessary to try a 1:1 and 1:2 ratio. My inserts are with sticky ends btw.
Thanks for your advice.