China
Africa
Explore chapters and articles related to this topic
Preformulation of New Biological Entities
Published in Sandeep Nema, John D. Ludwig, Parenteral Medications, 2019
Riccardo Torosantucci, Vasco Filipe, Jonathan Kingsbury, Atul Saluja, Yatin Gokarn
Characterization of the conformational stability of candidate biotherapeutics, as well as the effects of specific formulation variables on this stability, is generally studied by forced degradation vwof the molecule by chemical or thermal means. Chemical denaturation typically uses strong chaotropic agents such as urea or guanidine salts to induce loss of quaternary, tertiary, and secondary structures. Spectroscopic indicators of conformation are monitored as a function of denaturant concentration in order to generate an unfolding curve. This can be modeled from thermodynamic first principles to determine parameters related to conformational stability [126–128]. Alternatively, thermal denaturation typically employs either differential scanning calorimetry (DSC) or differential scanning fluorimetry (DSF). The former records the endothermic responses that accompany protein unfolding as the temperature is increased [129]. The resulting endotherms can be modeled from thermodynamic first principles [130]. The latter technique employs a dye responsive to the exposure of the hydrophobic interior of unfolded proteins during temperature ramping [131]. More recent technologies based on intrinsic fluorescence in heated capillaries have allowed greater throughput [132]. These technologies have been employed in the biopharmaceutical industry as relatively high-throughput and low-consumption tools for screening conformational stability [133–135]. Another useful approach to study the conformation of proteins is hydrogen/deuterium exchange mass spectrometry (H/DX MS). The different hydrogen/deuterium exchange rates of different hydrogen atoms at liable positions in proteins can provide conformation information at specific areas [136].
Bayesian model calibration for vacuum-ultraviolet photoionisation mass spectrometry
Published in Combustion Theory and Modelling, 2022
James Oreluk, Leonid Sheps, Habib Najm
Computational modelling of mass spectra has been investigated across several fields, with many works focusing on pre-processing methods and on identifying peaks in a noisy spectrum [9–11]. Nonparametric methods like wavelet regression are popular due to their flexibility and capacity to simultaneously denoise and model the background signal spectrum [12–15]. Unfortunately, these methods use basis functions and coefficients that are hard to interpret, making it difficult to obtain physically meaningful insight. An alternative physics-based parametric model approach can provide greater interpretability of the time-of-flight spectrum, albeit at the cost of reduced overall flexibility of the model. Several works have applied physics-based models to emulate specific mass spectrometry signals [16,17] within the constraints of the physics in their models. A study by Saltzberg et al. [16] used mass spectrometry data to update a physics-based model of the hydrogen/deuterium exchange process. Strubel et al. [17], applied Bayesian calibration to identify and estimate protein concentrations with uncertainties from liquid chromatography/mass spectrometry data.