Biacore X100 reference portal

Evaporation

Displacement of the solvent correction curve during the run could be caused by evaporation. Evaporation from vials results in increasing bulks while evaporation from running buffer results in decreasing bulks. Always use as full vials as possible to minimize the effects of evaporation. All vials should be capped immediately after preparation. The running buffer bottle should also be properly capped.

Evaporation from the solvent correction solutions shifts the solvent correction curve to the right.

Contamination

If one solution shows a drastic change in bulk it could be due to contamination from other solutions used in the instrument (for example wash solution). Contamination could be due to salt residues on the needle or if the distilled water bottle is empty. Accumulation of salt residues in the flow system and on the needle is prevented by following maintenance routines and by running standby between runs.

Contamination shifts single solvent correction points.

Unstable surface

The slope and shape of the curve depends on the difference in immobilized level between the reference spot and the immobilized spot. If ligand is lost from the immobilized spot, the difference between the surfaces is expected to decrease. Samples are always corrected using the closest correction curve before the sample. All changes in the surface between correction curves may affect the quality of the correction procedure. Correction curves should be run every 20 – 30 samples to assure valid correction of data.

Loss of ligand changes the slope of the curve.

Binding and precipitation

Irreversible binding or precipitation of samples on any spot will also change the correction curve. If a correction cycle is affected by for example precipitation it may be necessary to exclude the cycle. Samples affected by the precipitation should be re-tested. Using adaptive conditions for the samples can increase the stability and reduce effects on the correction curve.

Buffer refill

If buffer is added during the run there is a possibility that the difference between solvent correction solutions and the running buffer is altered. The effect can be seen as a change in the correction interval as well as a baseline shift. Always make sure if possible that there is sufficient buffer for the run. If buffer has to be refilled, make a note of the cycle where buffer was added. It will take a few cycles after addition for the new buffer to reach the flow cell.

Buffer changes can shift the curve.

System preparation

Correct system preparation is necessary to ensure that the first analysis cycles (including the first correction curve) are reliable. Correct system preparation includes:

• Change all buffers or refresh the flow system according to the requirements of the run
• Run sensor chip conditioning and start-up cycles where appropriate

Inadequate system preparation can displace the first solvent correction curve.

The solvent correction points do not fall on the fitted line

The quality of the corrected results will never be better than the quality of the correction curve. The quality of the correction curve is indicated mainly by the Chi2 value of the fit. The acceptable error is dependent on the response level of the sample but preferably the Chi2 values should not exceed 2 RU². Including blanks with different bulks is one way to test the quality of the correction procedure (corrected blank values should be close to zero regardless of the bulk contribution).

Old solvent correction solutions, a salt contaminated flow system and disturbed injections could all contribute to increased Chi2 values of the fit.

The first thing to do when the Chi2 values are high is to check the sensorgrams and exclude any cycles or correction points which are obviously disturbed.

Air injection during solvent correction cycle (left) gives high scattering in solvent correction plot (right).

If the sensorgrams are acceptable but the fit is still poor, there may be errors in preparation of the solvent correction solutions. The exact DMSO concentration in theses solutions is not important, but it is essential that the bulk response is matched as carefully as possible in all other respects. Variations in e.g. salt concentration will contribute to fitting errors. If one or two points deviate consistently from the fitted curve, these can be excluded from the fitting.

Bad points can be excluded from the fitting.

Sample report point fall outside the correction interval

Match samples, solvent correction solutions and running buffer carefully with respect to salt concentration. Samples with large bulk shifts indicate errors in sample concentration or buffer preparation procedure. Make sure that the solvent correction solutions used cover a correction range of about -500 to +1000 RU and that the negative control samples lie within this range

Note that the solvent correction range depends on the differences between the running buffer and the solvent correction solutions, not on absolute DMSO concentrations.

At 5% DMSO in standard running buffer, an error of 0.5% in the DMSO concentration introduces a bulk shift of about 800 RU during the sample injection. When samples are diluted from stock solutions in 100% DMSO, errors of this order of magnitude can easily be introduced.

Extend the range of solvent correction solutions if individual samples lie outside the range in spite of careful sample preparation. The extrapolation option in the software should be used carefully, since the correction curve is usually not linear and extrapolation can be misleading:

Illustration of the risks involved in extrapolating the solvent correction curve. Left: The correction curve extrapolated from the full set of measurements. Right: The same curve with the outermost points excluded from the fitting. The extrapolated curve is clearly incorrect.