Just to add my 2c worth... The department here has a couple of nanodrops as a shared facility, one for DNA/RNA and one for protein. It has been noticeable that over time people has been getting decreased reliability of measurements on the latter machine cf cuvette measurements, presumably due to the build-up of protein deposits over time - so I would say that although it's easier to clean than a cuvette, the nanodrop is not immune to the problem. The biggest issue I see with the nanodrop is evaporation of sample. Even here in moist Auckland, where RH is very often 80%+, taking a series of measurements with the nanodrop over a period of just a minute or two shows increasing concentration in the sample. So, for consistent results, one has to be careful to measure quickly. It's probably fine for comparative measurements, but as has been observed above, not great for super-accurate values for biophysics, and I think rather operator dependent. But all our students are super-careful, right? ;) Worth to note also that ProtParam calculates extinction coefficients based on Gill & von Hippel, (Gill, S.C. and von Hippel, P.H. (1989) Calculation of protein extinction coefficients from amino acid sequence data. Anal. Biochem. 182:319-326) who claim accuracy of ~5% for 'normal globular' proteins without extra chromophores. Whilst on this subject, I would put in a plug for the good old BCA (aka Pierce) assay for protein concentration. It's a little slow, but gets away from sequence dependency somewhat as it is primarily dependent on the peptide backbone rather than sidechains and works well in micro-titre plates etc. It is certainly very superior to Bradford. (Smith, P.K., et al. (1985). "Measurement of protein using bicinchoninic acid". Anal. Biochem. 150 (1): 76–85. doi:10.1016/0003-2697(85)90442-7).
cheers Shaun
