Applications and prospects of cryo-EM in drug discovery.

Drug discovery is a crucial part of human healthcare and has dramatically benefited human lifespan and life quality in recent centuries, however, it is usually time- and effort-consuming. Structural biology has been demonstrated as a powerful tool to accelerate drug development. Among different techniques, cryo-electron microscopy (cryo-EM) is emerging as the mainstream of structure determination of biomacromolecules in the past decade and has received increasing attention from the pharmaceutical industry. Although cryo-EM still has limitations in resolution, speed and throughput, a growing number of innovative drugs are being developed with the help of cryo-EM. Here, we aim to provide an overview of how cryo-EM techniques are applied to facilitate drug discovery. The development and typical workflow of cryo-EM technique will be briefly introduced, followed by its specific applications in structure-based drug design, fragment-based drug discovery, proteolysis targeting chimeras, antibody drug development and drug repurposing. Besides cryo-EM, drug discovery innovation usually involves other state-of-the-art techniques such as artificial intelligence (AI), which is increasingly active in diverse areas. The combination of cryo-EM and AI provides an opportunity to minimize limitations of cryo-EM such as automation, throughput and interpretation of medium-resolution maps, and tends to be the new direction of future development of cryo-EM. The rapid development of cryo-EM will make it as an indispensable part of modern drug discovery.

[The correlation based mid-infrared temperature and emissivity separation algorithm].

Temperature and emissivity separation is the key problem in infrared remote sensing. Based on the analysis of the relationship between the atmospheric downward radiance and surface emissivity containing atmosphere residue without the effects of sun irradiation, the present paper puts forward a temperature and emissivity separation algorithm for the ground-based mid-infrared hyperspectral data. The algorithm uses the correlation between the atmospheric downward radiance and surface emissivity containing atmosphere residue as a criterion to optimize the surface temperature, and the correlation between the atmospheric downward radiance and surface emissivity containing atmosphere residue depends on the bias between the estimated surface temperature and true surface temperature. The larger the temperature bias, the greater the correlation. Once we have obtained the surface temperature, the surface emissivity can be calculated easily. The accuracy of the algorithm was evaluated with the simulated mid-infrared hyperspectral data. The results of simulated calculation show that the algorithm can achieve higher accuracy of temperature and emissivity inversion, and also has broad applicability. Meanwhile, the algorithm is insensitive to the instrumental random noise and the change in atmospheric downward radiance during the field measurements.

[Multi-layer perceptron neural network based algorithm for simultaneous retrieving temperature and emissivity from hyperspectral FTIR data].

The present paper firstly points out the defect of typical temperature and emissivity separation algorithms when dealing with hyperspectral FTIR data: the conventional temperature and emissivity algorithms can not reproduce correct emissivity value when the difference between the ground-leaving radiance and object's blackbody radiation at its true temperature and the instrument random noise are on the same order, and this phenomenon is very prone to occur rence near 714 and 1 250 cm(-1) in the field measurements. In order to settle this defect, a three-layer perceptron neural network has been introduced into the simultaneous inversion of temperature and emissivity from hyperspectral FTIR data. The soil emissivity spectra from the ASTER spectral library were used to produce the training data, the soil emissivity spectra from the MODIS spectral library were used to produce the test data, and the result of network test shows the MLP is robust. Meanwhile, the ISSTES algorithm was used to retrieve the temperature and emissivity form the test data. By comparing the results of MLP and ISSTES, we found the MLP can overcome the disadvantage of typical temperature and emisivity separation, although the rmse of derived emissivity using MLP is lower than the ISSTES as a whole. Hence, the MLP can be regarded as a beneficial complementarity of the typical temperature and emissivity separation.