Multimodal Emotion Recognition Based on Facial Expressions, Speech, and EEG.

Goal: As an essential human-machine interactive task, emotion recognition has become an emerging area over the decades. Although previous attempts to classify emotions have achieved high performance, several challenges remain open: 1) How to effectively recognize emotions using different modalities remains challenging. 2) Due to the increasing amount of computing power required for deep learning, how to provide real-time detection and improve the robustness of deep neural networks is important. Method: In this paper, we propose a deep learning-based multimodal emotion recognition (MER) called Deep-Emotion, which can adaptively integrate the most discriminating features from facial expressions, speech, and electroencephalogram (EEG) to improve the performance of the MER. Specifically, the proposed Deep-Emotion framework consists of three branches, i.e., the facial branch, speech branch, and EEG branch. Correspondingly, the facial branch uses the improved GhostNet neural network proposed in this paper for feature extraction, which effectively alleviates the overfitting phenomenon in the training process and improves the classification accuracy compared with the original GhostNet network. For work on the speech branch, this paper proposes a lightweight fully convolutional neural network (LFCNN) for the efficient extraction of speech emotion features. Regarding the study of EEG branches, we proposed a tree-like LSTM (tLSTM) model capable of fusing multi-stage features for EEG emotion feature extraction. Finally, we adopted the strategy of decision-level fusion to integrate the recognition results of the above three modes, resulting in more comprehensive and accurate performance. Result and Conclusions: Extensive experiments on the CK+, EMO-DB, and MAHNOB-HCI datasets have demonstrated the advanced nature of the Deep-Emotion method proposed in this paper, as well as the feasibility and superiority of the MER approach.

Spherical cluster heterojunction engineering of NiFeP/g-C3N4 for efficient oxygen evolution reaction in alkaline solution.

The construction of heterojunctions can reduce the energy barrier for the oxygen evolution reaction (OER), which is crucial for the design of efficient electrocatalysts. A novel OER electrocatalyst, composed of g-C3N4-supported NiFeP spherical nanoclusters, was successfully synthesized using a simple hydrothermal method and a gas-phase precipitation method. Benefiting from its unique spherical nanocluster structure and strong electronic interactions among Ni, Fe, and P, the catalyst exhibited outstanding performance under alkaline conditions, with an overpotential of only 232 mV at a current density of 10 mA cm-2 and a Tafel slope of 103 mV dec-1. Additionally, the electrical resistance of NiFeP/g-C3N4 (Rct = 5.1 Ω) was much lower than that of NiFeP (Rct = 10.8 Ω) and layered g-C3N4 (Rct = 44.8 Ω). The formation of a Schottky barrier heterojunction efficiently reduced electron transfer impedance during the OER process, accelerating the electron transfer from g-C3N4 to NiFeP, enhancing the carrier concentration, and thereby improving the OER activity. Moreover, The robust g-C3N4 chain-mail protects NiFeP from adverse reaction environments, maintaining a balance between catalytic activity and stability. Furthermore, ab initio molecular dynamics (AIMD) and density functional theory (DFT) were conducted to explore the thermal stability and internal electron transfer behavior of the cluster heterojunction structure. This study offers a broader design strategy for the development of transition metal phosphide (TMPs) materials in the oxygen evolution reaction.

Effect of Annealing on Visible-Bubble Formation and Stability Profiles of Freeze-Dried High Concentration Omalizumab Formulations.

In the clinical application of freeze-dried highly concentrated omalizumab formulations, extensive visible bubbles (VBs) can be generated and remain for a long period of time in the reconstitution process, which greatly reduces the clinical use efficiency. It is necessary to understand the forming and breaking mechanism of VBs in the reconstitution process, which is a key factor for efficient and safe administration of biopharmaceutical injection. The effects of different thermal treatments on the volume of VBs and stability of omalizumab, mAb-1, and mAb-2 were investigated. The internal microvoids of the cake were characterized by scanning electron microscopy and mercury intrusion porosimetry. Electron paramagnetic resonance was applied to obtain the molecular mobility of the protein during annealing. A large number of VBs were generated in the reconstitution process of unannealed omalizumab and remained for a long period of time. When annealing steps were added, the volume of VBs was dramatically reduced. When annealed at an aggressive temperature (i.e., -6 °C), although the volume of VBs decreased, the aggregation and acidic species increased significantly. Thus, our observations highlight the importance of setting an additional annealing step with a suitable temperature, which contributes to reducing the VBs while maintaining the stability of the high concentration freeze-dried protein formulation.

FeCoNi molybdenum-based oxides for efficient electrocatalytic oxygen evolution reaction.

The search for highly efficient and inexpensive electrocatalysts is crucial to the advancement of environmentally friendly and sustainable energy sources. Here, adopting a one-step hydrothermal method, we have effectively fabricated a self-supported multi-metal molybdenum-based oxide (FeCoNi-MoO4) on nickel foam (NF). In addition to changing the catalyst's microstructure, the introducing of Fe and Co, enhanced its active center count, improved its electronic structure, and in turn reduced the difficulty for high-valence Ni and Fe species to form, which accelerates the oxygen evolution reaction (OER) kinetics by promoting the development of the actual active materials, NiOOH and FeOOH. FeCoNi-MoO4 has outstanding OER performance, requiring just 204 mV overpotentials at 10 mA cm-2 and 271 mV at 100 mA cm-2. Its exceptional OER kinetics at both low and high currents are indicated by a Tafel slope of 50.6 mV dec-1, which is attributed to the combined effect of its multi-metal composition and a higher number of active sites. Moreover, the FeCoNi-MoO4 electrode was operated continuously for over 48 h. Furthermore, the density functional theory (DFT) results demonstrated that the introducing of Fe and Co, which quickens the rate of electron transfer during the electrocatalytic process, improves the ability of oxygen intermediate species to adsorb, and ultimately lowers the overpotential, is responsible for the increased electrocatalytic activity of FeCoNi-MoO4. This work offers hope for further developments in the sector by proposing an efficient approach for creating multi-active electrocatalysts that are stable, economical, and efficient.

The Osmolality and Hemolysis of High-Concentration Monoclonal Antibody Formulations.

PURPOSES: Highly concentrated monoclonal antibody (mAb) formulations for subcutaneous administration are becoming increasingly preferred within the biopharmaceutical industry for ease of use and improved patient compliance. A common phenomenon observed in the industry is that osmolality detected via freezing-point depression (FPD) in high-concentration mAb formulations is much higher than the theoretical concentrations, yet the occurrence of this phenomenon and its possible safety issues have been rarely reported. METHODS: The current study summarized theoretical osmolality of U.S. Food and Drug Administration approved high-concentration mAb formulations and evaluated effects of high osmolality on safety using hemolysis experiments for the first time. Two mAbs formulated at 150 mg/mL were used as models and configured into two isotonic solutions: a, a theoretically calculated molarity in the isotonic range (H) and b, an osmolality value measured via the FPD in the isotonic range (I). The H and I formulations of each mAb were individually subjected to hemolysis experiments, and the hemolysis rates of the two formulations of the same mAb were compared. Besides, the effect of mAb concentration on osmolality detected by FPD was explored as well. RESULTS: The results indicated that the hemolysis rates were similar between the H and I formulations of mAbs at the same sample addition volume, and the osmolality values increased approximately linearly with the increase in mAb concentration. CONCLUSIONS: High osmolality for high-concentration mAb formulations would not affect product safety and the excipients could be added at relatively high levels to maintain product stability, especially for labile products.

The Effects of Excipients on Freeze-dried Monoclonal Antibody Formulation Degradation and Sub-Visible Particle Formation during Shaking.

PURPOSES: We previously reported an unexpected phenomenon that shaking stress could cause more protein degradation in freeze-dried monoclonal antibody (mAb) formulations than liquid ones (J Pharm Sci, 2022, 2134). The main purposes of the present study were to investigate the effects of shaking stress on protein degradation and sub-visible particle (SbVP) formation in freeze-dried mAb formulations, and to analyze the factors influencing protein degradation during production and transportation. METHODS: The aggregation behavior of mAb-X formulations during production and transportation was simulated by shaking at a rate of 300 rpm at 25°C for 24 h. The contents of particles and monomers were analyzed by micro-flow imaging, dynamic light scattering, size exclusion chromatography, and ultraviolet - visible (UV-Vis) spectroscopy to compare the protective effects of excipients on the aggregation of mAb-X. RESULTS: Shaking stress could cause protein degradation in freeze-dried mAb-X formulations, while surfactant, appropriate pH, polyol mannitol, and high protein concentration could impact SbVP generation. Water content had little effect on freeze-dried protein degradation during shaking, as far as the water content was controlled in the acceptable range as recommended by mainstream pharmacopoeias (i.e., less than 3%). CONCLUSIONS: Shaking stress can reduce the physical stability of freeze-dried mAb formulations, and the addition of surfactants, polyol mannitol, and a high protein concentration have protective effects against the degradation of model mAb formulations induced by shaking stress. The experimental results provide new insight for the development of freeze-dried mAb formulations.

Nutritional awareness of pregnant women and the underlying influencing factors.

Nutritional awareness is described as having knowledge or understanding of nutrition. It is often related to the ability of an individual to make an accurate estimate of their food intake, which involves comparing their actual nutritional behavior with the recommended food consumption. Nutritional awareness of women during the various phases of pregnancy may vary significantly across countries due to cultural and lifestyle differences. There has been extensive research on nutritional awareness of pregnant women in selected countries or regions; however, relatively few studies have explored it during different stages of pregnancy. To fill this gap, this article reviews the existing literature and draws together insights into the following areas: changes in nutritional awareness during various phases of pregnancy, nutritional awareness of pregnant women and its underlying factors in various nations, and the research methods used to study nutritional awareness of pregnant women.

Monitoring of low-molecular-weight protein aggregation by CE-SDS as a complementary method to SE-HPLC.

Capillary electrophoresis with sodium dodecyl sulfate (CE-SDS) has long been proven to have excellent performance in the analysis and characterization of therapeutic proteins. However, it is rarely used for the detection of low-molecular-weight proteins or peptides. Our research has proved the ability of CE-SDS to characterize the purity of low-molecular-weight proteins (i.e., <10 kDa) and even polypeptides. In this article, insulin glargine was used as a model protein, and CE-SDS was used to analyze the samples damaged by heating and light exposure. The monomers, dimers, and trimers of insulin glargine were effectively separated, and the results of the mass spectrometry also confirmed the existence of two kinds of insulin aggregates. For comparison, the size-exclusion high-performance liquid chromatography (SE-HPLC) only showed a single aggregate peak. In addition, the denaturation conditions caused only the covalent aggregates to appear in the CE-SDS analysis. These advantages also make CE-SDS an excellent supplementary technology to the traditional SE-HPLC, providing biopharmaceutical analysts with more information.

An Overview of the Stability and Delivery Challenges of Commercial Nucleic Acid Therapeutics.

Nucleic acid (NA)-based biopharmaceuticals have emerged as promising therapeutic modalities. NA therapeutics are a diverse class of RNA and DNA and include antisense oligonucleotides, siRNA, miRNA, mRNA, small activating RNA, and gene therapies. Meanwhile, NA therapeutics have posed significant stability and delivery challenges and are expensive. This article discusses the challenges and opportunities for achieving stable formulations of NAs with novel drug delivery systems (DDSs). Here we review the current progress in the stability issues and the significance of novel DDSs associated with NA-based biopharmaceuticals, as well as mRNA vaccines. We also highlight the European Medicines Agency (EMA) and US Food and Drug Administration (FDA)-approved NA-based therapeutics with their formulation profiles. NA therapeutics could impact future markets if the remaining challenges and requirements are addressed. Regardless of the limited information available for NA therapeutics, reviewing and collating the relevant facts and figures generates a precious resource for formulation experts familiar with the NA therapeutics' stability profile, their delivery challenges, and regulatory acceptance.

Characterization and exploration of an artifact in the reducing capillary electrophoresis-sodium dodecyl sulfate analysis of the 'me-too' drug zuberitamab related to rituximab.

For monoclonal antibody (mAb) drugs, the 'me-too' drug is a pharmacologically active compound that is structurally similar to the first-in-class drugs, acting on the same target and is used for the same therapeutic purposes, but it may differ in drug-drug interactions and adverse drug reactions. Capillary electrophoresis-sodium dodecyl sulfate (CE-SDS) has been widely used for quality evaluation of mAb drugs. The properties of the detected substances can interfere with the credibility and accuracy of the method. In the routine comparison analysis for both innovator rituximab and 'me-too' drug zuberitamab samples, an uncommon artifact related to the heavy chain (HC) of zuberitamab was observed in reducing CE-SDS and interfered with our identification of the purity of samples. In this work, the overall hydrophobicity of the HCs of rituximab, zuberitamab, and several other common mAbs was characterized and determined by reversed-phase high-performance liquid chromatography. Additionally, the local hydrophobicity and surface charge were compared using Expasy ProtScale and PyMOL software simulations. We concluded that noncovalent protein aggregation can be related to strong hydrophobicity and low electrostatic repulsion of local amino acid regions, which complicates drug quality control. These findings shed light on the relationship between protein aggregation and the local hydrophobicity region, and broaden the way to analyze the detection 'artifacts' in reducing CE-SDS studies of therapeutic proteins.

AGL-Net: An Efficient Neural Network for EEG-Based Driver Fatigue Detection.

BACKGROUND: In recent years, road traffic safety has become a prominent issue due to the worldwide proliferation of vehicles on roads. The challenge of driver fatigue detection involves balancing the efficiency and accuracy of the detection process. While various detection methods are available, electroencephalography (EEG) is considered the gold standard due to its high precision in terms of detecting fatigue. However, deep learning models for EEG-based fatigue detection are limited by their large numbers of parameters and low computational efficiency levels, making it difficult to implement them on mobile devices. METHODS: To overcome this challenge, an attention-based Ghost-LSTM neural network (AGL-Net) is proposed for EEG-based fatigue detection in this paper. AGL-Net utilizes an attention mechanism to focus on relevant features and incorporates Ghost bottlenecks to efficiently extract spatial EEG fatigue information. Temporal EEG fatigue features are extracted using a long short-term memory (LSTM) network. We establish two types of models: regression and classification models. In the regression model, we use linear regression to obtain regression values. In the classification model, we classify features based on the predicted values obtained from regression. RESULTS: AGL-Net exhibits improved computational efficiency and a more lightweight design than existing deep learning models, as evidenced by its floating-point operations per second (FLOPs) and Params values of 2.67 M and 103,530, respectively. Furthermore, AGL-Net achieves an average accuracy of approximately 87.3% and an average root mean square error (RMSE) of approximately 0.0864 with the Shanghai Jiao Tong University (SJTU) Emotion EEG Dataset (SEED)-VIG fatigued driving dataset, indicating its advanced performance capabilities. CONCLUSIONS: The experiments conducted with the SEED-VIG dataset demonstrate the feasibility and advanced performance of the proposed fatigue detection method. The effectiveness of each AGL-Net module is verified through thorough ablation experiments. Additionally, the implementation of the Ghost bottleneck module greatly enhances the computational efficiency of the model. Overall, the proposed method has higher accuracy and computational efficiency than prior fatigue detection methods, demonstrating its considerable practical application value.

Effects of polyol excipient stability during storage and use on the quality of biopharmaceutical formulations.

Biopharmaceuticals are formulated using a variety of excipients to maintain their storage stability. However, some excipients are prone to degradation during repeated use and/or improper storage, and the impurities generated by their degradation are easily overlooked by end users and are usually not strictly monitored, affecting the stability of biopharmaceuticals. In this study, we evaluated the degradation profile of polyol excipient glycerol during repeated use and improper storage and identified an unprecedented cyclic ketal impurity using gas chromatography with mass spectrometry (GC-MS). The other polyol excipient, mannitol, was much more stable than glycerol. The effects of degraded glycerol and mannitol on the stability of the model biopharmaceutical pentapeptide, thymopentin, were also evaluated. The thymopentin content was only 66.4% in the thymopentin formulations with degraded glycerol, compared to 95.8% in other formulations after the stress test. Most glycerol impurities (i.e., aldehydes and ketones) reacted with thymopentin, affecting the stability of thymopentin formulations. In conclusion, this work suggests that more attention should be paid to the quality changes of excipients during repeated use and storage. Additional testing of excipient stability under real or accelerated conditions by manufacturers would help avoid unexpected and painful results.

Current developments of bioanalytical sample preparation techniques in pharmaceuticals.

Sample preparation is considered as the bottleneck step in bioanalysis because each biological matrix has its own unique challenges and complexity. Competent sample preparation to extract the desired analytes and remove redundant components is a crucial step in each bioanalytical approach. The matrix effect is a key hurdle in bioanalytical sample preparation, which has gained extensive consideration. Novel sample preparation techniques have advantages over classical techniques in terms of accuracy, automation, ease of sample preparation, storage, and shipment and have become increasingly popular over the past decade. Our objective is to provide a broad outline of current developments in various bioanalytical sample preparation techniques in chromatographic and spectroscopic examinations. In addition, how these techniques have gained considerable attention over the past decade in bioanalytical research is mentioned with preferred examples. Modern trends in bioanalytical sample preparation techniques, including sorbent-based microextraction techniques, are primarily emphasized.

Investigation of an artifact during non-reduced capillary electrophoresis with sodium dodecyl sulfate analysis utilizing N-ethylmaleimide as an alkylation reagent.

This manuscript describes the formation of an artifact shoulder peak with a slightly larger retention time than the main peak under the standard non-reduced capillary electrophoresis with sodium dodecyl sulfate (nrCE-SDS) analysis of a therapeutic recombinant protein X, and clarifies the formation mechanism of the artifact caused by N-ethylmaleimide (NEM) during the sample preparation procedure. A design of experiment (DoE) approach was used to investigate the impact of the factors on the formation of the impurity. Additionally, orthogonal analytical experiments were performed to study the root cause of this phenomenon. The results consistently suggested that the Michael addition reaction between NEM and lysine residues in protein X, and decreased electrophoretic mobility due to increased molecular weight, was the root cause for the artifact, which could be partially inhibited by modifications of incubation conditions. Thus, before performing the nrCE-SDS method, the effects of alkylation reagents and sample preparation procedure on analytical results need to be considered seriously.

Elimination of light chain tailing in reducing capillary electrophoresis with sodium dodecyl sulfate analysis of a monoclonal antibody.

Capillary electrophoresis with sodium dodecyl sulfate (CE-SDS) is a common analytical technique for investigating the purity and molecular size heterogeneity of monoclonal antibody (mAb) drugs. In reducing CE-SDS analysis of mAb-A, the light chain (LC) peak exhibited severe tailing, seriously affecting the purity analysis. The purposes of this investigation are to clarify the source of tailing and develop a more appropriate CE-SDS method to eliminate LC tailing. The degree of LC tailing was closely related to the mAb concentration, SDS concentration, and injection amount, and more hydrophobic detergents, such as sodium hexadecyl sulfate (SHS) and sodium tetradecyl sulfate (STS), could be used instead of SDS to obtain better peak shapes. The results also indicated that the tailing was caused by the binding problem associated with SDS, and SHS/STS could provide a more stable and uniform complexation for the LC. In summary, the method we developed successfully eliminated the LC tailing and provided a robust characterization of mAb-A in reducing CE-SDS analysis.

Identification and characterization of chemical and physical stability of insulin formulations utilizing degraded glycerol after repeated use and storage.

Insulin treatment is currently considered to be the main strategy for controlling diabetes. Although the recombinant insulin formulation is relatively mature, we found that a batch of insulin formulation exhibited an unusual degradation rate in the stability experiment. The main purposes of this article are to identify the root cause for this phenomenon and characterize of chemical and physical degradation products. We compared the chemical and physical stability of two batches of insulin formulations prepared separately with simulated repeated use and freshly opened glycerol. The chemical stability of insulin was identified by liquid chromatography coupled with tandem mass spectrometry (LC- MS/MS). Micro-flow imaging (MFI), far-ultraviolet circular dichroism (Far-UV CD) and Thioflavin T (ThT) fluorescent assays were used to reveal protein aggregation and fibrosis. The chemical and physical stability of the insulin formulation with newly opened glycerol was much better than that with degraded glycerol, and both groups of formulations were extremely sensitive to light. The results indicated that the original batch insulin formulation with abnormal stability was indeed caused by the excipient glycerol after long-term storage and repeated usage. More attention should be paid to the quality changes of excipients during repeated usage and storage of excipients for the practical purpose. Moreover, we have discovered a novel degradation pathway for insulin and peptides in general. In addition, LC-MS/MS results suggested that the N-terminus of insulin B-chain was prone to chemical degradation which enlightens that it could be potentially modified to improve the stability of insulin formulations.

Investigation of an Uncommon Artifact during Reducing Capillary Electrophoresis-Sodium Dodecyl Sulfate Analysis of a Monoclonal Antibody with Dynamic Light Scattering and Reversed Phase High-Performance Liquid Chromatography.

PURPOSES: In reducing capillary electrophoresis sodium dodecyl sulfate (CE-SDS) analysis of a monoclonal antibody (mAb-1), the peak area ratio of heavy chain (HC) to light chain (LC) was out of balance, while multiple artifact peaks were observed following the migration of HC. The main purposes of this study were to describe the techniques utilized to eliminate this artifact and clarify the root cause for this interesting phenomenon. METHODS: We optimized the CE-SDS analysis of mAb-1 by a vairety of techniques including changing the concentration of protein or replacing SDS with a more hydrophobic surfactant (i.e., sodium hexadecyl sulfate (SHS) or sodium tetradecyl sulfate (STS) instead of SDS) in sample and/or the sieving gel buffer. Dynamic light scattering (DLS) and reversed phase high-performance liquid chromatography (RP-HPLC) were used to study the protein-surfactant complex. RESULTS: The artifact could be partially mitigated by reducing the protein concentration and replacing SDS with SHS or STS in the sample and/or the sieving gel buffer solutions. Due to replacing a more hydrophobic surfactant, the HC-surfactant complex formed was more resistant to dissociation, preventing additional hydrophobic HC-HC interaction and aggregation, thus eliminating the artifact problem. CONCLUSIONS: DLS and RP-HPLC are powerful supplementary techniques in characterizing the protein-surfactant complex, and hydrophobic surfactants such as SHS and STS could afford more normal electropherograms during the analysis of mAbs.

Freeze-Dried Biopharmaceutical Formulations are Surprisingly Less Stable than Liquid Formulations during Dropping.

PURPOSES: This article describes an interesting phenomenon in which optimized freeze-dried (FD) biopharmaceutical formulations are generally more prone to degradation than their liquid counterparts during dropping and proposes an underlying cause for this surprising phenomenon. METHODS: Two monoclonal antibodies (mAbs) and a fusion protein (FP) were used as model biopharmaceuticals. The stability after dropping stress was determined by ultraviolet-visible (UV-Vis), size exclusion high-performance liquid chromatography (SE-HPLC), micro-flow imaging (MFI), and dynamic light scattering (DLS). RESULTS: Contrary to what we would normally assume, the FD formulations of the three biopharmaceuticals studied here generally showed much higher amounts of protein sub-visible particles (SbVPs) than liquid formulations after applying the same dropping stress as determined by MFI and DLS. Traditional techniques, such as UV-Vis and SE-HPLC, could hardly detect such degradation. CONCLUSIONS: We propose that the higher temperature caused by dropping for the FD powders than the liquid formulations was probably one of the root causes for the higher amount of particles formed for the FD powders. We also recommend that dropping stress should be included for early-stage screening and choosing liquid versus FD biopharmaceutical formulations.

Freeze-Dried Monoclonal Antibody Formulations are Unexpectedly More Prone to Degradation Than Liquid Formulations Under Shaking Stress.

Liquid biopharmaceuticals including monoclonal antibodies (mAbs) have been widely acknowledged to undergo various stresses during shipping/handling and long-term storage. Several mechanical stresses including shaking during shipping has been widely known to cause protein aggregation and sub-visible particle (SbVP) formation in liquid biopharmaceutical formulations. However, shaking-induced degradation of freeze-dried (FD) biopharmaceuticals has seldomly been reported in the literature and therefore this type of stress is widely overlooked in industry due to their presumed high stability, especially when the formulations and freeze-drying processes are fully optimized. In this Lessons Learned article, we report an interesting phenomenon in which the optimized FD biopharmaceutical formulations of three typical mAbs showed much degradation and SbVP formation under shaking stress compared with their liquid counterparts. This is a striking deviation to the notion that mAbs are generally more stable in the FD formulations than in the liquid ones. Therefore, shaking stress experiment should be considered a critical stress condition for early-stage selection of liquid versus FD mAb formulations.

Characterization of Grinding-Induced Subvisible Particles and Free Radicals in a Freeze-Dried Monoclonal Antibody Formulation.

PURPOSES: The primary objectives of this study were to investigate the degradation mechanisms of freeze-dried monoclonal antibody (mAb) formulations under mechanical grinding, assess the sensitivity and suitability of various particle analysis techniques, analyze the structure of the collected subvisible particles (SbVPs), and analyze the antioxidant mechanism of methionine (Met) under degradation process to gain a thorough understanding of the phenomenon. METHODS: The freeze-dried mAb-X formulations underwent grinding, and the resultant SbVPs were characterized through visual inspection, flow imaging microscopy, dynamic light scattering, ultraviolet-visible spectroscopy, and size-exclusion high-performance liquid chromatography. We further evaluated the effect of different temperatures and the free radical scavenger Met on SbVP formation. The produced free radicals were detected using electron paramagnetic resonance, and Met S-oxide formation was detected using liquid chromatography-mass spectrometry. In addition, we analyzed the obtained SbVPs using capillary electrophoresis sodium dodecyl sulfate and Fourier transform infrared spectroscopy. RESULTS: Grinding leads to SbVP formation under high temperature and free radical formation. Free radicals produced during grinding require the participation of a macromolecule. Met could then bind to the produced free radicals, thus partially protecting mAb-X from degradation while itself undergoing oxidation to form Met(O). Sensitivity differences between different particle analysis techniques were evaluated, and the obtained SbVPs showed significant changes in secondary structure and the formation of covalent aggregates and fragments. CONCLUSIONS: Met plays the role of an antioxidant in protecting macromolecules by quenching the free radicals produced during grinding. To thoroughly characterize SbVPs, multiple and orthogonal particle analysis techniques should be used, and if necessary, SbVPs should be processed by enrichment to accurately analyze primary and high order structures.