Intensity of Specific Vs. Non-specific binding

Hello everybody and Happy New Year!

I'm making my own SPR (LSPR based) handheld machine. (I know, weird)
Want to try to use it for diagnostic purposes - so need to be sure that it is worth to buy antigens and antibodies.
While feeling comfortable with device and sensor making, have lack of knowledge in bio.
Please advise, is my thinking correct or wrong?

I'm using Glycerol in DI solutions for calibration. Can see jumps from 0.05% to 0.1% of Glycerol in DI.
Calculations says that it corresponds to LOD ~60RU. Worse than commercial SPR machine.
Using bare gold (no surface chemistry) can see curves down to 10nM of BSA in DI (corresponds to ~66ng/ml).
Conventional rapid tests have LOD ~1ng/ml.

In my thinking:
Many people use BSA in buffers for SPR. And still have signals for analyte-ligand binding.
With surface chemistry BSA non-specifically binds more than just to bare gold.
If non-specific binding of BSA on bare gold at 10nM can be seen, then specific binding will give more signal.

So this is a question:
can it be estimated that if
it is possible to see curve for non-specific binding at level of ~60ng/ml,
then
it will be possible to see specific binding at level of ~1ng/ml?

Thank you for reading!
Doiiu

Hi Doiiu,
Most specific measurements are done with sensor chips where the bare gold is covered with a pacifying layer (e.g. dextran) to lower non-specific binding. And the pacifying layer is also modified to make a specific layer to detect. ( www.sprpages.nl/sensor-chips-intro )
In addition most SPR responses are measured from a base line. Say that your flow buffer with BSA has a response of 100 RU then the specific response will be specific RU minus 100 RU.

You could try to contact someone at a close by university and ask for some antigen/antibody. Maybe they can give you something for free.

Kind regards
Arnoud