Not sure if it fits this forum but I thought I will try, maybe someone tackled this issue before.
Been using OT-2 for 3 years for ELISA in 96 and more recently 384 plates. There is a “zig-zag” pattern in all direct ELISA EC50 experiments where antibody was captured on 384 high binding plate (100ng/well, working volume 25ul). Meaning, the 2nd, 4th (and sometimes 6th) concentrations have higher signal than their respective previous conc. point - and this is consistent over many experiments. If I flip the format (analyte captured on plate and antibody in serial dilution) this phenomenon disappears.
Wow this is like the first deep experimental troubleshooting I’ve seen on here.
So I notice the “staggered offsets” are actually still occurring even at the 7th and 8th data points. The left point isn’t higher than the right point, but it’s still higher than you’d expect given the curve, and the right point is lower than you’d expect.
Here are some thoughts/ suggestions:
1.Try it manually and see if you reproduce the same error.
2. Look at the physical layout of sample wells on the plate. There might be something revealing there, like the even numbered points are always on a certain side. Incubators can have spatial temperature gradients and this can definitely impact sample incubations. Offsets on the OT-2 might be impacting a certain pattern of wells.
3. Just watch what’s happening up close to make sure you’re seeing the volumes and heights and precisions you’d expect to. Maybe run a water test with dye and do an absorption assay to check that you’re getting the right volumes/ dilutions in the wells.
I think the first two points are the most revealing. The first dilution has a higher signal than an undiluted sample, like what? It seems unlikely to have to do with the liquid handling itself based on this. You’re not “undiluting” your sample, my guess is it’s an artifact of some other phenomenon.
Hey @Stefan, thank you for the input.
I did try to perform the dilutions with a dye and it came out perfect, <5% error between duplicates and between dilution points.
I will need to try doing the experiment manually but if the dye test was successful I feel like that’s not where the issue stems from. My initial idea was to change plate source, but the fact that when I flip the assay and introduce avidity (my antibodies are symmetrical) I don’t see this issue doesn’t give me much hope.