Grams ai software

FT-IR spectroscopy was carried out by following the method described previously [31 (link), 34 (link)] with some modifications. For FT-IR measurement, 50 μl of ~7.5 μg of purified recombinant protein sample (full length or deletion versions of Pol λ) was taken in a microcon filter device fitted with a 3-kDa cut-off membrane. The protein sample was diluted with ~200 μl D₂O (Sigma). The sample was then centrifuged at 14000 rpm for 8 to 10 min at 4°C until the sample volume reached ~50 μl. It was further diluted with ~200 μl D₂O and again centrifuged. This D₂O exchange process was repeated 3–4 times. Finally, the D₂O exchanged protein sample (10 μl) was carefully placed in between two clean CaF₂ windows separated by a 50 μm thick teflon spacer. FT-IR scans were collected in the range of 1600–1700 cm^-1 at the resolution of 2 cm^-1 using a Spectrum 100 FT-IR spectrometer (Perkin Elmer). Spectrum of D₂O containing buffer was subtracted from each sample spectrum. Fourier self-deconvolution was used to resolve the peak positions in the broad amide I contour. Curve fitting of the original amide I contours was performed using Thermo GRAMS AI software [31 (link), 34 (link)]. The percentage of each secondary structure was calculated by dividing the area of the corresponding peak with the total area of the amide I contour.

All Raman measurements
were conducted
with a WITec Alpha300 Access confocal Raman microscope using a 633
nm laser with a power of 1–5 mW, 20× or 50× magnification
air objectives, and a built-in CCD camera for obtaining the bright
field images (WITec). TEM grids and loose MP powders were supported
on clean glass coverslips to ensure that the samples were not lost.
Raw spectral data were extracted from the accompanying WITec Control
5 software so that cosmic ray removal (zap) and baseline corrections
(multi-point baseline subtraction) could be conducted using Grams
AI software (version 9.3; ThermoFisher Scientific). The final presented
spectra were obtained by accumulating multiple (300–1000) 0.5
s measurements and averaging them. A full table of the exact measurement
parameters for each of the presented spectra can be seen in Table S3.

CD spectra were recorded with application of a Chirascan Plus Spectrometer (Applied Photophysics, U.K.). The purified stalk dimers and their truncated variants were dialyzed against 50 mM Tris-HCl pH 7.4, 150 mM NaCl, 10 mM MgCl₂. The protein concentrations were determined from the absorbance at 280 nm using an extinction coefficient for each protein complex. Spectra were recorded in the range 200–250 nm, at 25 °C temperature, with a 1 nm resolution. The scan rate was 60 nm/min. Protein samples for CD were scanned three times and averaged. The averaged baseline spectra were then subtracted from the averaged sample spectra and converted to molar ellipticity. The recorded spectra were analyzed with application of the Grams/AI software from Thermo Scientific (USA).

Micro-Raman resonance (RR) spectra were acquired using a Jasco TRS 300 triple monochromator spectrometer (2400 lines/mm grid) (Jasco Europe, Cremella, Italy) equipped with an Andor CCD detector (Andor Technology Ltd., Belfast, UK) and interfaced with an Olympus BH-2 microscope (Olympus, Tokyo, Japan), provided with three different objectives (10×, 20× and 50×). A Cobolt Twist TM 25 laser (Cobolt, Stockholm, Sweden) with emission at 457 nm and maximum power of 25 mW was used as excitation source. For each analysis the spectral region 1000–1800 cm⁻¹ was considered and the spectra were acquired as the sum of 60 accumulations with an exposure time of 2 seconds.
The Raman spectra were corrected using the Savitsky-Golay smoothing algorithm and the baseline correction by means of the Grams/AI software (Thermo Fisher Scientific, Waltham, MA, USA).
The Fourier-transform (FT) Raman spectrum of delphinidin 3,5-diglucoside chloride was acquired using a Jasco RFT-600 spectrophotometer (Jasco Europe, Cremella, Italy), equipped with a Nd-YAG laser (1064 nm). The output laser power was 100 mW. The spectrum was recorded as sum of 300 accumulations with resolution 4 cm⁻¹.

Peptides of interest were identified manually by searching their m/z-values within the experimental mass spectrum. For the quantification, specific ion current (SIC) chromatograms of peptides of interest were generated on the basis of their monoisotopic mass and detected charge states using GRAMS AI software (Thermo Fisher Scientific, Dreieich, Germany). Relative amounts of Asn deamidation, Asp isomerization, and Met oxidation were calculated by manual integration of modified and unmodified peptide peaks. The level of afucosylation was calculated by manual integration of afucosylated and fucosylated glycan species.

The FT-NIR spectra of the mock-up samples and pure binders were processed by principal component analysis (PCA), performed by the MINITAB 14 software. The spectra were first truncated between 6000 and 4000 cm⁻¹ and normalized between 0 and 1. Subsequently, they were transformed into the corresponding 1st derivatives using GRAMS AI software (Thermo Fisher Scientific) and the Savitsky–Golay algorithm (2nd degree polynomial and 31 points), to eliminate the contribution of the baseline slope and to emphasize the differences among the spectra. The variables to which PCA analysis is applied are, therefore, the dA/d $\overline{\nu }$ values in the range 6000–4000 cm⁻¹. The covariance matrix was used to give less weight to baseline points in the calculation.
To calculate the ratios between band intensities, the baseline was subtracted from the spectrum and the band heights were evaluated by using the Grams/AI software. Absorption bands typical of oil or egg yolk, respectively, were chosen to exploit the ratio as discriminating value between the two binders (see Section 3.3 for more details). When bands due to the gypsum ground layer were observed in the NIR spectrum besides those of the binder, the contribution of calcium sulphate dihydrate was subtracted using the same software.