Pharmaceutical analysis is a discipline based on chemical, physical, biological, and information technologies. At present, biotechnological analysis is a short branch in pharmaceutical analysis; however, bioanalysis is the basis and an important part of medicine. Biotechnological approaches can provide information on biological activity and even clinical efficacy and safety, which are important characteristics of drug quality. Because of their advantages in reflecting the overall biological effects or functions of drugs and providing visual and intuitive results, some biotechnological analysis methods have been gradually applied to pharmaceutical analysis from raw material to manufacturing and final product analysis, including DNA super-barcoding, DNA-based rapid detection, multiplex ligation-dependent probe amplification, hyperspectral imaging combined with artificial intelligence, 3D biologically printed organoids, omics-based artificial intelligence, microfluidic chips, organ-on-a-chip, signal transduction pathway-related reporter gene assays, and the zebrafish thrombosis model. The applications of these emerging biotechniques in pharmaceutical analysis have been discussed in this review.

Alzheimer's disease (AD) is gradually increasing in prevalence and the complexity of its pathogenesis has led to a lengthy process of developing therapeutic drugs with limited success. Faced with this challenge, we proposed using a state-of-the-art drug screening algorithm to identify potential therapeutic compounds for AD from traditional Chinese medicine formulas with strong empirical support. We developed four deep neural network (DNN) models for AD drugs screening at the disease and target levels. The AD model was trained with compounds labeled for AD activity to predict active compounds at the disease level, while the acetylcholinesterase (AChE), monoamine oxidase-A (MAO-A), and 5-hydroxytryptamine 6 (5-HT₆) models were trained for specific AD targets. All four models performed excellently and were used to identify potential AD agents in the Kaixinsan (KXS) formula. High-scoring compounds underwent experimental validation at the enzyme, cellular, and animal levels. Compounds like 2,4-di-tert-butylphenol and elemicin showed significant binding and inhibitory effects on AChE and MAO-A. Additionally, 13 compounds, including α-asarone, penetrated the blood-brain barrier (BBB), indicating potential brain target binding, and eight compounds enhanced microglial β-amyloid phagocytosis, aiding in clearing AD pathological substances. Our results demonstrate the effectiveness of deep learning models in developing AD therapies and provide a strong platform for AD drug discovery.

A dual-column tandem mode was used to establish the fingerprints of Sarcandrae herb from different origins, and their chemical compositions were characterized by UPLC-Q-TOF-MS/MS, which provided an experimental basis for the establishment of a rapid and efficient method for the overall quality control of Sarcandrae herba. For the first time, nine common components were identified from the Sarcandrae herba herbs of 24 origins, which were neochlorogenic acid, chlorogenic acid, 4-caffeoylquinic acid, eleutheroside B1, quercetin-3-O-β-D-glucuronide, neoastilbin, astilbin, isofraxidin, and rosmarinic acid, respectively. A total of 92 compounds were identified by liquid mass spectrometry. The quality of the Sarcandrae herb from 24 origins was analyzed by similarity evaluation, principal component analysis, and cluster analysis, and the chemical components of Sarcandrae herba were identified by UPLC-Q-TOF-MS/MS. The results showed that the overall analysis based on fingerprinting and mass spectrometry could differentiate the origins of the herbs.

Background: Tablet manufacturing is challenged by strong dynamic coupling of process parameters, significant material property fluctuations, and delayed quality control, with tablet weight stability being particularly critical in high-speed production. Traditional static optimization methods relying on empirical judgment struggle to manage these dynamics, leading to substantial variations in tablet weight and hardness that severely compromise production efficiency. Methods: This study proposes a data-driven closed-loop control system centered on a novel Iterative Learning Model Predictive Control (IL-MPC) architecture. The core innovation lies in directly integrating iterative learning constraints within the MPC optimization framework. This constraint-embedding mechanism enables systematic utilization of historical batch data while preserving real-time optimization capabilities. The IL-MPC approach achieves enhanced batch-to-batch performance consistency with reduced computational burden, effectively combining the dual advantages of learning and optimization. Results: Simulation experiments and industrial production data validate the practical feasibility of the IL-MPC algorithm. Implementation results demonstrate that the proposed system effectively manages dynamic process variations, significantly improving control precision for both tablet weight and hardness, outperforming conventional control methods. Conclusions: This research breaks through the technical bottleneck of dynamic regulation in tablet manufacturing. The developed IL-MPC framework provides a reproducible closed-loop control paradigm for intelligent pharmaceutical manufacturing, promoting the industry’s transformation toward data-driven models and advancing intelligent drug production.