Bioinformatics and biomedical informatics with ChatGPT: Year one review.

The year 2023 marked a significant surge in the exploration of applying large language model chatbots, notably Chat Generative Pre-trained Transformer (ChatGPT), across various disciplines. We surveyed the application of ChatGPT in bioinformatics and biomedical informatics throughout the year, covering omics, genetics, biomedical text mining, drug discovery, biomedical image understanding, bioinformatics programming, and bioinformatics education. Our survey delineates the current strengths and limitations of this chatbot in bioinformatics and offers insights into potential avenues for future developments.

Green and efficient catalytic oxidation of ethylbenzene to acetophenone over cobalt oxide supported on a carbon material derived from sugar.

The use of biomass as a carbon source to support metal oxides has significant advantages in environmental protection and reducing the cost of the catalyst. The critical point lies on the development of highly active and recyclable catalysts. In this study, sugar was used as a carbon source, and cobalt oxide was loaded via an in situ method and an impregnation method to prepare the catalyst, respectively. The catalysts were characterized by scanning electron microscopy (SEM), transmission electron microscopy (TEM), high-resolution transmission electron microscopy (HRTEM), X-ray powder diffraction (XRD), X-ray photoelectron spectroscopy (XPS), Fourier transform infrared (FT-IR) spectroscopy, Raman, in situ electron paramagnetic resonance (in situ EPR) and N2 adsorption-desorption. The characterization results show that the macroporous carbon-supported cobalt oxide catalyst prepared by the in situ method contains CoOx nanoparticles on the support, but the dispersion of cobalt oxide on the support is more uniform compared with the catalyst prepared by the impregnation method. The catalytic performance of the prepared catalyst was evaluated through the oxidation of ethylbenzene (EB) to acetophenone (AP) as a probe reaction. Under the optimized reaction conditions, temperature (T) = 80 °C, m(catalyst) : m(EB) = 0.15, n(H2O2) : n(EB) : n(KBr) = 14.4 : 1 : 0.1, t = 8 h, and V(EB) : V(CH3COOH) = 1 : 10, the conversion of EB and the selectivity of AP were 84.1% and 81.3%, respectively. CoOx/SC-10-in situ exhibits improved reactivity of EB oxidation owing to cobalt ions on the carbon support that promote the free radical and acid catalytic reaction pathway. The cobalt oxide catalyst supported by biomass carbon has decent recycling and regeneration ability, which provides a new idea for the application of biomass.

Strengthening China's National Essential Public Health Services Package for hypertension and diabetes care: protocol for an interrupted time series study with mixed-methods process evaluation and health economic evaluation.

BACKGROUND: Despite major primary health care (PHC) reforms in China with the 2009 launch of the National Essential Public Health Service Package, the country experiences many challenges in improving the management of non-communicable diseases in PHC facilities. "EMERALD" is a multifaceted implementation strategy to strengthen the management of hypertension and type-2 diabetes mellitus (T2DM) in PHC facilities. The study aims to: (1) examine the effectiveness of EMERALD in improving hypertension and T2DM management; (2) evaluate the implementation of the interventions; and (3) use the study findings to model the long-term health economic impact of the interventions. METHODS: The EMERALD intervention components include: (1) empowerment for PHC providers through training and capacity building; (2) empowerment for patient communities through multi-media health education; and  (3) empowerment for local health administrators through health data monitoring and strengthening governance of local PHC programs. An interrupted time series design will be used to determine the effectiveness of the interventions based on routinely collected health data extracted from local health information systems. The primary effectiveness outcome is the guideline-recommended treatment rates for people with hypertension and T2DM. Secondary effectiveness outcomes include hypertension and T2DM diagnosis and control rates, and enrolment and adherence rates to the recommended care processes in the National Essential Public Health Service Package. A mixed-methods process evaluation will be conducted to evaluate the implementation of the interventions, including the reach of the target population, adequacy of adoption, level of implementation fidelity, and maintenance. Qualitative interviews with policy makers, health administrators, PHC providers, and patients with hypertension and/or T2DM will be conducted to further identify factors influencing the implementation. In addition, health economic modelling will be performed to explore the long-term incremental costs and benefits of the interventions. DISCUSSION: This study is expected to generate important evidence on the effectiveness, implementation, and health economic impact of complex PHC interventions to strengthen the primary care sector's contribution to addressing the growing burden of non-communicable diseases in China. TRIAL REGISTRATION: The study has been registered on Chinese Clinical Trial Registry at https://www.chictr.org.cn/ (Registration number ChiCTR2400082036, on March 19th 2024).

Co-designing interventions to strengthen the primary health care system for the management of hypertension and type 2 diabetes in China.

Background: Policy makers and researchers are tasked with exploring ways to strengthen primary health care (PHC) to address the growing burden of non-communicable diseases (NCDs). This study aims to use a co-design approach (i.e., meaningful involvement of research end users in study planning and design) to develop PHC interventions to improve the management of hypertension and type 2 diabetes (T2DM) in four study sites in China. Methods: The study adopted a three-step co-design approach, including (1) a two-round Delphi panel with health system and NCD professionals to identify prioritised health system challenges, (2) three co-design workshops (in each study site) with local health administrators, PHC providers, and residents with hypertension and/or T2DM, respectively, to develop interventions and identify factors influencing implementation, and (3) another round of co-design workshops with local health administrators to summarise findings and reach consensus. Qualitative synthesis was conducted to analyse results from the workshops. Findings: Thirteen experts were involved in the two-round Delphi panel, which identified three prioritised health system challenges, including limited capacities of PHC providers, suboptimal service quality and evaluation mechanisms, and unreliable health information systems. The co-design workshops involved 116 local stakeholders in 16 sessions (four in each site), and developed three groups of interventions to address the challenges: (1) empowering PHC providers through on-the-job training for capacity building; (2) empowering patient communities through health education on healthy lifestyles and NCD self-management; and (3) empowering health administrators through local health data monitoring and strengthening governance for local PHC programs. Site-specific interventions were also considered to cater for different local contexts. Several recommendations were further identified for the implementation of these interventions, emphasising the importance of local customisation, community participation, and cross-sectoral collaborations. Interpretation: By engaging multiple stakeholders in priority setting and solution generation, this study summarised several key areas for change in health workforce, service delivery, and health information. Future research should examine the effectiveness and implementation of these interventions to improve NCD management in PHC in China. Funding: This study is funded by National Health and Medical Research Council (NHMRC) Global Alliance for Chronic Diseases funding (APP1169757) and National Natural Science Foundation of China (72074065). Shangzhi Xiong is supported by University of New South Wales tuition fee scholarship.

An Efficient Muscle Segmentation Method Via Bayesian Fusion of Probabilistic Shape Modeling and Deep Edge Detection.

OBJECTIVE: Paraspinal muscle segmentation and reconstruction from MR images are critical to implement quantitative assessment of chronic and recurrent low back pains. Due to unclear muscle boundaries and shape variations, current segmentation methods demonstrate suboptimal performance with insufficient training samples. This work proposes a novel approach to modeling and inferring muscle shapes that enhances segmentation accuracy and efficiency with few training data. METHODS: Firstly, a probabilistic shape model (PSM) based on Fourier basis functions and Gaussian processes (GPs) is designed to encode 3D muscle shapes, where anatomical meanings are attributed to the model's geometric parameters. Muscle shape variations and correlations are described by the GPs of the geometric parameters, which allow a small size of parameters to model the distribution of muscle shapes. Secondly, a Bayesian framework is developed to achieve entire muscle segmentation by posterior estimations. The framework fuses the geometric prior of the PSM with observations of deep-learning-based edge detections (DED) and sparse manual annotations, by which issues of unclear boundaries and shape variations can be compensated. RESULTS AND CONCLUSION: Experiments on public and clinical datasets demonstrate that, with just three manually annotated slices, our method achieves a Dice similarity coefficient exceeding 90%, which outperforms other methods. Meanwhile, our method needs only a small training dataset and offers rapid inference speeds in clinical applications. SIGNIFICANCE: Our study enables precise assessment of paraspinal muscles in 2D and 3D, aiding clinicians and researchers in understanding muscle changes in various conditions, potentially enhancing treatment outcomes.

Bioinformatics and biomedical informatics with ChatGPT: Year one review.

The year 2023 marked a significant surge in the exploration of applying large language model (LLM) chatbots, notably ChatGPT, across various disciplines. We surveyed the applications of ChatGPT in bioinformatics and biomedical informatics throughout the year, covering omics, genetics, biomedical text mining, drug discovery, biomedical image understanding, bioinformatics programming, and bioinformatics education. Our survey delineates the current strengths and limitations of this chatbot in bioinformatics and offers insights into potential avenues for future developments.

FSP1-mediated ferroptosis in cancer: from mechanisms to therapeutic applications.

Ferroptosis is a new discovered regulated cell death triggered by the ferrous ion (Fe2+)-dependent accumulation of lipid peroxides associated with cancer and many other diseases. The mechanism of ferroptosis includes oxidation systems (such as enzymatic oxidation and free radical oxidation) and antioxidant systems (such as GSH/GPX4, CoQ10/FSP1, BH4/GCH1 and VKORC1L1/VK). Among them, ferroptosis suppressor protein 1 (FSP1), as a crucial regulatory factor in the antioxidant system, has shown a crucial role in ferroptosis. FSP1 has been well validated to ferroptosis in three ways, and a variety of intracellular factors and drug molecules can alleviate ferroptosis via FSP1, which has been demonstrated to alter the sensitivity and effectiveness of cancer therapies, including chemotherapy, radiotherapy, targeted therapy and immunotherapy. This review aims to provide important frameworks that, bring the regulation of FSP1 mediated ferroptosis into cancer therapies on the basis of existing studies.

Scientific figures interpreted by ChatGPT: strengths in plot recognition and limits in color perception.

Emerging studies underscore the promising capabilities of large language model-based chatbots in conducting basic bioinformatics data analyses. The recent feature of accepting image inputs by ChatGPT, also known as GPT-4V(ision), motivated us to explore its efficacy in deciphering bioinformatics scientific figures. Our evaluation with examples in cancer research, including sequencing data analysis, multimodal network-based drug repositioning, and tumor clonal evolution, revealed that ChatGPT can proficiently explain different plot types and apply biological knowledge to enrich interpretations. However, it struggled to provide accurate interpretations when color perception and quantitative analysis of visual elements were involved. Furthermore, while the chatbot can draft figure legends and summarize findings from the figures, stringent proofreading is imperative to ensure the accuracy and reliability of the content.

A Fish-like Binocular Vision System for Underwater Perception of Robotic Fish.

Biological fish exhibit a remarkably broad-spectrum visual perception capability. Inspired by the eye arrangement of biological fish, we design a fish-like binocular vision system, thereby endowing underwater bionic robots with an exceptionally broad visual perception capacity. Firstly, based on the design principles of binocular visual field overlap and tangency to streamlined shapes, a fish-like vision system is developed for underwater robots, enabling wide-field underwater perception without a waterproof cover. Secondly, addressing the significant distortion and parallax of the vision system, a visual field stitching algorithm is proposed to merge the binocular fields of view and obtain a complete perception image. Thirdly, an orientation alignment method is proposed that draws scales for yaw and pitch angles in the stitched images to provide a reference for the orientation of objects of interest within the field of view. Finally, underwater experiments evaluate the perception capabilities of the fish-like vision system, confirming the effectiveness of the visual field stitching algorithm and the orientation alignment method. The results show that the constructed vision system, when used underwater, achieves a horizontal field of view of 306.56°. The conducted work advances the visual perception capabilities of underwater robots and presents a novel approach to and insight for fish-inspired visual systems.

A critical period for developing face recognition.

Face learning has important critical periods during development. However, the computational mechanisms of critical periods remain unknown. Here, we conducted a series of in silico experiments and showed that, similar to humans, deep artificial neural networks exhibited critical periods during which a stimulus deficit could impair the development of face learning. Face learning could only be restored when providing information within the critical period, whereas, outside of the critical period, the model could not incorporate new information anymore. We further provided a full computational account by learning rate and demonstrated an alternative approach by knowledge distillation and attention transfer to partially recover the model outside of the critical period. We finally showed that model performance and recovery were associated with identity-selective units and the correspondence with the primate visual systems. Our present study not only reveals computational mechanisms underlying face learning but also points to strategies to restore impaired face learning.

TS-1@MCM-48 Core-Shell Catalysts for Efficient Oxidation of p-Diethylbenzene to High Value-Added Derivatives.

To expand the market capacity of p-diethylbenzene (PDEB), core-shell zeolite (TS-1@MCM-48) is designed as a catalyst for PDEB oxidation. TS-1@MCM-48 catalyst is synthesized by in-situ crystallization method and characterized by X-ray diffraction (XRD), transmission electron microscope (TEM), scanning electron microscope (SEM), X-ray photoelectron spectroscopy (XPS), N2 adsorption-desorption, in-situ electron paramagnetic resonance (EPR) and 29Si nuclear magnetic resonance (29Si MAS-NMR). Oxidation of PDEB by H2O2 was investigated systematically in liquid phase. The conversion of PDEB over TS-1@MCM-48 was 28.1 % and the total selectivity was 72.6 %, where the selectivity of EAP (p-ethylacetophenone) and EPEA (4-ethyl-α-methylbenzyl alcohol) was 28.6 % and 44.0 %, respectively. Compared with TS-1 and MCM-48 zeolite, the conversion rate of reactants and the selectivity of products have been significantly improved. The catalytic performance of TS-1@MCM-48 is derived from its well-crystallized microporous core and mesoporous shell with regular channels, which make active sites of TS-1 zeolite in the catalyst be fully utilized and mass transfer resistance be largely reduced. Further through theoretical calculation, we propose that the oxidation of PDEB is the result of the combination and mutual transformation of free radical process and carbocation process. Core-shell structure ensures the conversion rate of raw materials and improves the selectivity of products.

Neural mechanisms of face familiarity and learning in the human amygdala and hippocampus.

Recognizing familiar faces and learning new faces play an important role in social cognition. However, the underlying neural computational mechanisms remain unclear. Here, we record from single neurons in the human amygdala and hippocampus and find a greater neuronal representational distance between pairs of familiar faces than unfamiliar faces, suggesting that neural representations for familiar faces are more distinct. Representational distance increases with exposures to the same identity, suggesting that neural face representations are sharpened with learning and familiarization. Furthermore, representational distance is positively correlated with visual dissimilarity between faces, and exposure to visually similar faces increases representational distance, thus sharpening neural representations. Finally, we construct a computational model that demonstrates an increase in the representational distance of artificial units with training. Together, our results suggest that the neuronal population geometry, quantified by the representational distance, encodes face familiarity, similarity, and learning, forming the basis of face recognition and memory.

Ultrasensitive Ratiometric Electrochemiluminescence Sensor with an Efficient Antifouling and Antibacterial Interface of PSBMA@SiO2-MXene for Oxytetracycline Trace Detection in the Marine Environment.

The sensitivity and accuracy of electrochemiluminescence (ECL) sensors for detecting small-molecule pollutants in environmental water are affected not only by nonspecific adsorption of proteins and other molecules but also by bacterial interference. Therefore, there is an urgent need to develop an ECL sensor with antifouling and antibacterial functions for water environment monitoring. Herein, a highly efficient antifouling sensing interface (PSBMA@SiO2-MXene) based on zwitterionic sulfobetaine methacrylate (SBMA) antifouling nanospheres (NPs) and two-dimensional MXene nanosheets was designed for the sensitive detection of oxytetracycline (OTC), an antibiotic small-molecule pollutant. Specifically, SBMA with good hydrophilicity and electrical neutrality was connected to SiO2 NPs, thus effectively reducing protein and bacterial adsorption and improving stability. Second, MXene with a high specific surface area was selected as the carrier to load more antifouling NPs, which greatly improves the antifouling performance. Meanwhile, the introduction of MXene also enhances the conductivity of the antifouling interface. In addition, a ratio-based sensing strategy was designed to further improve the detection accuracy and sensitivity of the sensor by utilizing Au@luminol as an internal standard factor. Based on antifouling and antibacterial interfaces, as well as internal standard and ratiometric sensing strategies, the detection range of the proposed sensor was 0.1 ng/mL to 100 µg/mL, with a detection limit of 0.023 ng/mL, achieving trace dynamic monitoring of antibiotics in complex aqueous media.

TMEM147 aggravates the progression of HCC by modulating cholesterol homeostasis, suppressing ferroptosis, and promoting the M2 polarization of tumor-associated macrophages.

BACKGROUND: The endoplasmic reticulum (ER) regulates critical processes, including lipid synthesis, which are affected by transmembrane proteins localized in the ER membrane. One such protein, transmembrane protein 147 (TMEM147), has recently been implicated for its role in hepatocellular carcinoma (HCC) tumorigenesis; however, the mechanisms remain unclear. We investigated the role of TMEM147 in HCC and the underlying mechanisms. METHODS: TMEM147 expression was examined in human HCC cells and adjacent non-tumorous tissues using quantitative reverse transcription-polymerase chain reaction, western blotting, and immunohistochemistry. In vitro and in vivo studies were conducted to investigate the impact of TMEM147 on the progression of HCC. Proteins interacting with TMEM147 were identified via RNA-seq, immunoprecipitation, and mass spectrometry analyses. Lipidomic analysis and enzyme-linked immunosorbent assay (ELISA) were employed to determine and analyze cholesterol and 27-hydroxycholesterol (27HC) contents. Extensive experimental techniques were used to study ferroptosis in HCC cells. The fatty acid content of macrophages affected by TMEM147 was quantified using ELISA. Macrophage phenotypes were determined using immunofluorescence assay and flow cytometric analysis. RESULTS: TMEM147 mRNA and protein levels were increased in HCC cells, and the increased TMEM147 expression was associated with a poor survival. TMEM147 promoted tumor cell proliferation and metastases in vitro and in vivo. The protein was found to interact with the key enzyme 7-dehydrocholesterol reductase (DHCR7), which affected cellular cholesterol homeostasis and increased the extracellular levels of 27HC in HCC cells. TMEM147 also promoted the expression of DHCR7 by enhancing the activity of signal transducer and activator of transcription 2. 27HC expression upregulated glutathione peroxidase 4 in HCC, leading to ferroptosis resistance and promotion of HCC proliferation. HCC cell-derived 27HC expression increased the lipid metabolism in macrophages and activated peroxisome proliferator-activated receptor-γ signaling, thereby activating M2 macrophage polarization and promoting HCC cell invasion and migration. CONCLUSIONS: Our results indicate that TMEM147 confers ferroptosis resistance and M2 macrophage polarization, which are primarily dependent on the upregulation of cellular cholesterol homeostasis and 27HC secretion, leading to cancer growth and metastasis. These findings suggest that the TMEM147/STAT2/DHCR7/27HC axis in the tumor microenvironment may serve as a promising therapeutic target for HCC.

Bioinformatics Illustrations Decoded by ChatGPT: The Good, The Bad, and The Ugly.

Emerging studies underscore the promising capabilities of large language model-based chatbots in conducting fundamental bioinformatics data analyses. The recent feature of accepting image-inputs by ChatGPT motivated us to explore its efficacy in deciphering bioinformatics illustrations. Our evaluation with examples in cancer research, including sequencing data analysis, multimodal network-based drug repositioning, and tumor clonal evolution, revealed that ChatGPT can proficiently explain different plot types and apply biological knowledge to enrich interpretations. However, it struggled to provide accurate interpretations when quantitative analysis of visual elements was involved. Furthermore, while the chatbot can draft figure legends and summarize findings from the figures, stringent proofreading is imperative to ensure the accuracy and reliability of the content.

Three anthocyanin-rich berry extracts regulate the in vitro digestibility of corn starch: Physicochemical properties, structure and α-amylase.

This study aimed to compare the regulatory effects of blue honeysuckle anthocyanins (BHA), blueberry anthocyanins (BBA), and blackcurrant anthocyanins (BCA) on the in vitro digestibility of corn starch in terms of starch physicochemical properties and structure, as well as α-amylase inhibition. The results revealed that adding all three anthocyanins lowered digestibility in the following order: BHA > BCA > BBA. The terminal digestibility (C∞) decreased from 73.84 % to 57.3 % with the addition of 10 % BHA, while the resistant starch (RS) content increased from 4.39 % to 48.82 %. Fourier transform infrared spectroscopy (FTIR) and X-ray diffraction (XRD) analysis indicated that anthocyanins and starch interacted through noncovalent bonds. Differential scanning calorimetry (DSC) analysis showed that the gelatinization enthalpy was dramatically lowered in all three anthocyanin groups, with 10 % BHA producing a 38.58 % drop. Rheological property analysis showed that anthocyanins increased the apparent viscosity and modulus with starch. The interaction between anthocyanin and α-amylase was mainly through the formation of hydrogen bonds and hydrophobic forces. This research provides theoretical guidance for developing low glycemic index (GI) anthocyanin starch-based foods.

A three-in-one nanoplatform with photo/chemodynamic activities combined with glutathione depletion for treating bacterial infections.

The misuse of antibiotics leading to bacterial multidrug resistance is responsible for severe infectious diseases and a significant cause of mortality worldwide, resulting in numerous human disasters. Photodynamic antibacterial therapy (PDAT) is a promising strategy against multiantibiotic-resistant bacteria, but its antibacterial activity is greatly limited by reduced glutathione (GSH) in bacteria. In this study, we constructed a nanoplatform through the formation of metal chelating complexes (FeP) between ferric and pyrophosphate ions, with subsequent adsorption of the photosensitizer ZnPc(COOH)8 (octa-carboxyl substituted zinc phthalocyanine) mediated by polylysine (PL) on the surface. The nanocomplexes FeP@PL:ZnPc(COOH)8 exhibited the capacity of GSH depletion and chemodynamic activity, which synergistically enhanced PDAT efficacy. FeP@PL:ZnPc(COOH)8 possessed the excellent antibacterial activity in vivo and in vitro, which might be attributed to the increased production of reactive oxygen species (ROS), reduced GSH level in bacteria, improved bacterial uptake, and enhanced destruction of the bacterial outer membrane. Moreover, FeP@PL:ZnPc(COOH)8 exhibited accelerated wound healing efficacy and the ability to recognize bacteria-infected wounds, rendering it an effective theranostic nanoplatform for bacterial infections. The construction strategy of nanocomplexes in this study holds theoretical and practical significance for high-efficiency synergistic photodynamic and chemodynamic antibacterial therapy.

Construction of chitosan-based thermosensitive composite hydrogels for recognizing and combined chemo-photodynamic elimination of Gram-negative bacterial infections.

Recently, rapid acquisition of bacterial resistance and consequent slow healing of infected wounds threaten human life and health. In this study, chitosan-based hydrogels and nanocomplexes ZnPc(COOH)8:PMB composed of photosensitizer ZnPc(COOH)8 and antibiotic polymyxin B (PMB) were integrated into a thermosensitive antibacterial platform ZnPc(COOH)8:PMB@gel. Interestingly, fluorescence and reactive oxygen species (ROS) of ZnPc(COOH)8:PMB@gel can be triggered by E. coli bacteria at 37 °C, but not by S. aureus bacteria, which gave the potential to simultaneously detect and treat Gram-negative bacteria. The survival rate for a certain amount of E. coli bacteria treated with ZnPc(COOH)8:PMB (ZnPc(COOH)8 2 µM) was decreased by approximately fivefold than that with either ZnPc(COOH)8 or PMB alone, indicating combined antibacterial efficacy. ZnPc(COOH)8:PMB@gel facilitated the complete healing of wounds infected with E. coli bacteria in about seven days, while over 10 % wounds treated with ZnPc(COOH)8 or PMB remained unhealed on the 9th day. ZnPc(COOH)8:PMB resulted in a threefold increase of ZnPc(COOH)8 fluorescence in E. coli bacteria suggesting enhanced uptake of ZnPc(COOH)8 for the intervention of PMB on membrane permeability. The construction principle of the thermosensitive antibacterial platform and the combined antimicrobial strategy can be applied to other photosensitizers and antibiotics for detection and treatment of wound infections.

Generating minimum set of gRNA to cover multiple targets in multiple genomes with MINORg.

MINORg is an offline gRNA design tool that generates the smallest possible combination of gRNA capable of covering all desired targets in multiple non-reference genomes. As interest in pangenomic research grows, so does the workload required for large screens in multiple individuals. MINORg aims to lessen this workload by capitalising on sequence homology to favour multi-target gRNA while simultaneously screening multiple genetic backgrounds in order to generate reusable gRNA panels. We demonstrated the practical application of MINORg by knocking out 11 homologous genes tandemly arrayed in a multi-gene cluster in two Arabidopsis thaliana lineages using three gRNA output by MINORg. We also described a new PCR-free modular cloning system for multiplexing gRNA, and used it to knockout three tandemly arrayed genes in another multi-gene cluster with gRNA designed by MINORg. Source code is freely available at https://github.com/rlrq/MINORg.