Deep learning and computer vision techniques for microcirculation analysis: A review

Mesophilic Argonaute-Mediated Polydisperse Droplet Biosensor for Amplification-Free, One-Pot, and Multiplexed Nucleic Acid Detection Using Deep Learning

Detection of nucleic acids from a single multiplexed and amplification-free test is critical for ensuring food safety, clinical diagnostics, and environmental monitoring. In this study, we introduced a mesophilic Argonaute protein from Clostridium butyricum (CbAgo), which exhibits nucleic acid endonuclease activity, to achieve a programmable, amplification-free system (PASS) for rapid nucleic acid quantification at ambient temperatures in one pot. By using CbAgo-mediated binding with specific guide DNA (gDNA) and subsequent targeted cleavage of wild-type target DNAs complementary to gDNA, PASS can detect multiple foodborne pathogen DNA (<102 CFU/mL) simultaneously. The fluorescence signals were then transferred to polydisperse emulsions and analyzed by using deep learning. This simplifies the process and increases the suitability of polydisperse emulsions compared to traditional digital PCR, which requires homogeneous droplets for accurate detection. We believe that PASS has the potential to become a next-generation point-of-care digital nucleic acid detection method.

Quantitative structure-activity relationship study of amide derivatives as xanthine oxidase inhibitors using machine learning

The target of the study is to predict the inhibitory effect of amide derivatives on xanthine oxidase (XO) by building several models, which are based on the theory of the quantitative structure-activity relationship (QSAR). The heuristic method (HM) was used to linearly select descriptors and build a linear model. XGBoost was used to non-linearly select descriptors, and radial basis kernel function support vector regression (RBF SVR), polynomial kernel function SVR (poly SVR), linear kernel function SVR (linear SVR), mix-kernel function SVR (MIX SVR), and random forest (RF) were adopted to establish non-linear models, in which the MIX-SVR method gives the best result. The kernel function of MIX SVR has strong abilities of learning and generalization of established models simultaneously, which is because it is a combination of the linear kernel function, the radial basis kernel function, and the polynomial kernel function. In order to test the robustness of the models, leave-one-out cross validation (LOOCV) was adopted. In a training set, R2 = 0.97 and RMSE = 0.01; in a test set, R2 = 0.95, RMSE = 0.01, and Rcv2 = 0.96. This result is in line with the experimental expectations, which indicate that the MIX-SVR modeling approach has good applications in the study of amide derivatives.