I run indirect ELISAs for COVID antibodies. That's my whole job right now as a research technician. I coat the plates with a pure RBD, then detect antibodies in serum. Recently it's come to light that I've run plates with false positives. Running those same serum samples again gives VERY different (negative) results. If entire plates were giving false positives we could figure out what mistakes I'm making, but for the most part the duplicates look good, the standard is good, the dilutions are good... The person who wrote this protocol hasn't included a negative control because there have been occasions during the development of the protocol where pooled serum from 2018 or earlier was used as a negative, but those pooled serum controls were somehow consistently more positive than that of several confirmed, post-recovery COVID patients. Obviously that lack of negative control has predictably come back to bite us.

Originally we thought it was a plate washer malfunction, but it turns out these smatterings of false positives predates the plate washer acting up.

The short of the long is that some of my plates have very clear, obvious negatives, some borderline positives, and some clear, obvious positives. The issue is that some of those positives are false positives and neither I nor my supervisor can for the life of us figure out why.

Ideas would be awesome. ELISAs are my entire job right now, and the PI is understandably not thrilled with me.

Additional details: Plates are washed by an automatic plate washer after the incubation of sample dilutions, and after the incubation of the secondary antibody/HRP. The plates are blocked before sample addition with a fractionated bovine serum/gelatin/tween20 blocking buffer. Dilutions are made in that same blocking buffer, and the secondary antibody is also diluted in blocking buffer. We're using Immulon 4HBX plates. The plates are developed with ABTS in McIlvain's plus H2o2.

Ask my anything else that might help figure this out. I'll answer what I can.


2 Answers 2


Is your detection or secondary antibody aggregating? I've done extensive clinical ELISA development in the past and when this problem inevitably crops up for these assays the most likely culprits in order were:

  1. Plate washer acting funny
  2. Antibody aggregates binding nonspecifically to the plate
  3. Inefficient blocking reagent/not enough time spent blocking
  4. Person running the samples is mixing things up

Even if your false positives predate the platewasher malfunction, you could still have a leaky/clogged dispensing nozzle. Try rotating the plate 180 between wash steps or hand-washing and see if that helps.

For aggregated secondary/detection ab, give them a brief spin and pipette off the top, and for blocking increase volume/time.



Quite possibly the samples change over time. The antibodies your test selects for may be reacting in the messy biological sample.

If it were possible, grab the people giving the interesting samples and get a PCR test on them to see if they are positive or not.

An experiment on the effect of waiting time on the samples would be interesting, it might be nice if it were performed at another lab to reduce your stress.

I hope your discovery gets you cited in relevant research or industry publications.

  • $\begingroup$ Thank you! We collect saliva as well with each visit to check for active infection. I'll bring up the idea about the antibodies reacting in the sera and waiting time. This is my first tech position post-undergrad, and I love it very much so I want to do the best I can. Thank you again! $\endgroup$ Dec 23, 2020 at 16:04

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .