To align the spectra, a mass shift of <0.1% was determined.
In the case of the amide-II peak near 1,550 cm−1, we also observed no consistent variation for the γ-FBS samples after aligning the spectra with the baseline at 900 cm−1 for comparison (data not shown).
Concerning the amide-II peak and after aligning the spectra at 900 cm−1 for comparison, there was also a general increase in absorbance with γ-irradiation but, again, this increase did not correlate strictly with the level of γ-dose used (Fig. 4B).
To obtain the same number of frames, a dynamic time warping method is used to align the extracted features (spectrum and F0) of the source voice and target voice.
Dataset was elaborated to align and normalize the spectra.
Vertical lines between prophages shows aligned regions whose hues of grey correspond to the percentage identity of aligned regions as specified in the spectrum at the bottom right of each image.
In the aligned spectrum, there are two additional peaks due to the scattering from Al and O in the amorphous Al2O3 surface oxide (typically approximately 4 nm thick), which formed upon exposure of the sample to air.
When the MS spectra are aligned with the MS spectrum, the corroborating b-ions match 'vertically' while the matching y-ions are found at different positions depending on the parent mass of each MS spectrum (Fig. 2).
The spectrum of an aligned sample of E7 alone was also recorded versus air at each of these orientations, and subtractions of these reference spectra were used to obtain the data presented.
While many existing 1D alignment methods have attempted to align each spectrum in a data set to a single consensus spectrum, the quality of these "star alignments" can suffer if there is wide chemical diversity between spectra, as any individual spectrum chosen as a star may not include the full complement of metabolites present in the other data sets.
