Although the approach with firefly luciferases seems to be straightforward, a drawback exists: A spectrum of firefly luciferase is known to change in a pH-dependent manner, which may hamper spectral deconvolution or precise photon counts with specific filters.
Having their advantages in being transparent and computationally straightforward, the main drawbacks of I/O models lie in the focus on short-term effects and the non-capability of feedback effects [87].
After the other non-silver elements are evaporated and carried away by the carrier gas, leaving only the silver particles in the furnace, the evaporation/condensation technique is easy and straightforward; however, the major drawback of this technique is large energy consumption and slow process.
This fit provides a straightforward handle on several drawbacks of the technology: saturation (by reconstructing the missing signal beyond the saturation limit), overshining (by reconstructing the signal below the influenced feature) and spotting variability (by automatically adjusting the integration range to the source size).
The only drawback to its straightforward presentation is that if the hand doing the seasoning happens to lay it on a little heavy, there's not much to mitigate it.
However, such straightforward approach would have the drawback of being strongly biased towards what is known, whereas the goal of unsupervised machine learning is to uncover what is not.
The main drawback of using straightforward clustering of mass spectra is that it is negatively affected by the pixel-to-pixel variability issue (Alexandrov et al., 2010; Watrous et al., 2011), which is especially serious in the most popular IMS techniques, SIMS and MALDI-IMS.
While this formulation is straightforward, it bears some major drawbacks preventing computability for practical scenarios.
A straightforward approach to eliminate these drawbacks is to use multivariate polynomial interpolation [ 13] for the construction of the functions (see Methods).
