GPT-4 can pass the Korean National Licensing Examination for Korean Medicine Doctors.

Traditional Korean medicine (TKM) emphasizes individualized diagnosis and treatment. This uniqueness makes AI modeling difficult due to limited data and implicit processes. Large language models (LLMs) have demonstrated impressive medical inference, even without advanced training in medical texts. This study assessed the capabilities of GPT-4 in TKM, using the Korean National Licensing Examination for Korean Medicine Doctors (K-NLEKMD) as a benchmark. The K-NLEKMD, administered by a national organization, encompasses 12 major subjects in TKM. GPT-4 answered 340 questions from the 2022 K-NLEKMD. We optimized prompts with Chinese-term annotation, English translation for questions and instruction, exam-optimized instruction, and self-consistency. GPT-4 with optimized prompts achieved 66.18% accuracy, surpassing both the examination's average pass mark of 60% and the 40% minimum for each subject. The gradual introduction of language-related prompts and prompting techniques enhanced the accuracy from 51.82% to its maximum accuracy. GPT-4 showed low accuracy in subjects including public health & medicine-related law, internal medicine (2), and acupuncture medicine which are highly localized in Korea and TKM. The model's accuracy was lower for questions requiring TKM-specialized knowledge than those that did not. It exhibited higher accuracy in diagnosis-based and recall-based questions than in intervention-based questions. A significant positive correlation was observed between the consistency and accuracy of GPT-4's responses. This study unveils both the potential and challenges of applying LLMs to TKM. These findings underline the potential of LLMs like GPT-4 in culturally adapted medicine, especially TKM, for tasks such as clinical assistance, medical education, and research. But they also point towards the necessity for the development of methods to mitigate cultural bias inherent in large language models and validate their efficacy in real-world clinical settings.

Predicting activatory and inhibitory drug-target interactions based on structural compound representations and genetically perturbed transcriptomes.

A computational approach to identifying drug-target interactions (DTIs) is a credible strategy for accelerating drug development and understanding the mechanisms of action of small molecules. However, current methods to predict DTIs have mainly focused on identifying simple interactions, requiring further experiments to understand mechanism of drug. Here, we propose AI-DTI, a novel method that predicts activatory and inhibitory DTIs by combining the mol2vec and genetically perturbed transcriptomes. We trained the model on large-scale DTIs with MoA and found that our model outperformed a previous model that predicted activatory and inhibitory DTIs. Data augmentation of target feature vectors enabled the model to predict DTIs for a wide druggable targets. Our method achieved substantial performance in an independent dataset where the target was unseen in the training set and a high-throughput screening dataset where positive and negative samples were explicitly defined. Also, our method successfully rediscovered approximately half of the DTIs for drugs used in the treatment of COVID-19. These results indicate that AI-DTI is a practically useful tool for guiding drug discovery processes and generating plausible hypotheses that can reveal unknown mechanisms of drug action.

Identifying Candidate Flavonoids for Non-Alcoholic Fatty Liver Disease by Network-Based Strategy.

Nonalcoholic fatty liver disease (NAFLD) is the most common type of chronic liver disease and lacks guaranteed pharmacological therapeutic options. In this study, we applied a network-based framework for comprehensively identifying candidate flavonoids for the prevention and/or treatment of NAFLD. Flavonoid-target interaction information was obtained from combining experimentally validated data and results obtained using a recently developed machine-learning model, AI-DTI. Flavonoids were then prioritized by calculating the network proximity between flavonoid targets and NAFLD-associated proteins. The preventive effects of the candidate flavonoids were evaluated using FFA-induced hepatic steatosis in HepG2 and AML12 cells. We reconstructed the flavonoid-target network and found that the number of re-covered compound-target interactions was significantly higher than the chance level. Proximity scores have successfully rediscovered flavonoids and their potential mechanisms that are reported to have therapeutic effects on NAFLD. Finally, we revealed that discovered candidates, particularly glycitin, significantly attenuated lipid accumulation and moderately inhibited intracellular reactive oxygen species production. We further confirmed the affinity of glycitin with the predicted target using molecular docking and found that glycitin targets are closely related to several proteins involved in lipid metabolism, inflammatory responses, and oxidative stress. The predicted network-level effects were validated at the levels of mRNA. In summary, our study offers and validates network-based methods for the identification of candidate flavonoids for NAFLD.

Investigating the Biomarkers of the Sasang Constitution via Network Pharmacology Approach.

Sasang constitutional (SC) medicine classifies people into Soeum (SE), Soyang (SY), Taeeum (TE), and Taeyang (TY) types based on psychological and physical traits. However, biomarkers of these types are still unclear. We aimed to identify biomarkers among the SC types using network pharmacology methods. Target genes associated with the SC types were identified by grouping herb targets that preserve and strengthen the requisite energy (Bomyeongjiju). The herb targets were obtained by constructing an herb-compound-target network. We identified 371, 185, 146, and 89 target genes and their unique biological processes related to SE, SY, TE, and TY types, respectively. While the targets of SE and SY types were the most similar among the target pairs of the SC types, those of TY type overlapped with only a few other SC-type targets. Moreover, SE, SY, TE, and TY were related to "diseases of the digestive system," "diseases of the nervous system," "endocrine, nutritional, and metabolic diseases," and "congenital malformations, deformations, and chromosomal abnormalities," respectively. We successfully identified the target genes, biological processes, and diseases related to each SC type. We also demonstrated that a drug-centric approach using network pharmacology analysis provides a deeper understanding of the concept of Sasang constitutional medicine at a phenotypic level.

Machine learning-based prediction of Sasang constitution types using comprehensive clinical information and identification of key features for diagnosis.

BACKGROUND: Despite the importance of accurate Sasang type diagnosis, a unique form of Korean medicine, there have been concerns about consistency among diagnoses. We investigate a data-driven integrative diagnostic model by applying machine learning to a multicenter clinical dataset with comprehensive features. METHODS: Extremely randomized trees (ERT), support vector machines, multinomial logistic regression, and K-nearest neighbor were applied, and performances were evaluated by cross-validation. The feature importance of the classifier was analyzed to understand which information is crucial in diagnosis. RESULTS: The ERT classifier showed the highest performance, with an overall f1 score of 0.60 ± 0.060. The feature classes of body measurement, personality, general information, and cold-heat were more decisive than others in classifying Sasang types. Costal angle was the most informative feature. In pairwise classification, we found Sasang type-dependent distinctions that body measurement features played a key role in TE-SE and TE-SY datasets, while personality and cold-heat features showed importance in SE-SY dataset. CONCLUSION: Current study investigated a comprehensive diagnostic model for Sasang type using machine learning and achieved better performance than previous studies. This study helps data-driven decision making in clinics by revealing key features contributing to the Sasang type diagnosis.

A Novel Framework for Understanding the Pattern Identification of Traditional Asian Medicine From the Machine Learning Perspective.

Pattern identification (PI), a unique diagnostic system of traditional Asian medicine, is the process of inferring the pathological nature or location of lesions based on observed symptoms. Despite its critical role in theory and practice, the information processing principles underlying PI systems are generally unclear. We present a novel framework for comprehending the PI system from a machine learning perspective. After a brief introduction to the dimensionality of the data, we propose that the PI system can be modeled as a dimensionality reduction process and discuss analytical issues that can be addressed using our framework. Our framework promotes a new approach in understanding the underlying mechanisms of the PI process with strong mathematical tools, thereby enriching the explanatory theories of traditional Asian medicine.

The Methodological Trends of Traditional Herbal Medicine Employing Network Pharmacology.

Natural products, including traditional herbal medicine (THM), are known to exert their therapeutic effects by acting on multiple targets, so researchers have employed network pharmacology methods to decipher the potential mechanisms of THM. To conduct THM-network pharmacology (THM-NP) studies, researchers have employed different tools and databases for constructing and analyzing herb-compound-target networks. In this study, we attempted to capture the methodological trends in THM-NP research. We identified the tools and databases employed to conduct THM-NP studies and visualized their combinatorial patterns. We also constructed co-author and affiliation networks to further understand how the methodologies are employed among researchers. The results showed that the number of THM-NP studies and employed databases/tools have been dramatically increased in the last decade, and there are characteristic patterns in combining methods of each analysis step in THM-NP studies. Overall, the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP) was the most frequently employed network pharmacology database in THM-NP studies. Among the processes involved in THM-NP research, the methodology for constructing a compound-target network has shown the greatest change over time. In summary, our analysis describes comprehensive methodological trends and current ideas in research design for network pharmacology researchers.

Systems-level mechanisms of action of Panax ginseng: a network pharmacological approach.

Panax ginseng has been used since ancient times based on the traditional Asian medicine theory and clinical experiences, and currently, is one of the most popular herbs in the world. To date, most of the studies concerning P. ginseng have focused on specific mechanisms of action of individual constituents. However, in spite of many studies on the molecular mechanisms of P. ginseng, it still remains unclear how multiple active ingredients of P. ginseng interact with multiple targets simultaneously, giving the multidimensional effects on various conditions and diseases. In order to decipher the systems-level mechanism of multiple ingredients of P. ginseng, a novel approach is needed beyond conventional reductive analysis. We aim to review the systems-level mechanism of P. ginseng by adopting novel analytical framework-network pharmacology. Here, we constructed a compound-target network of P. ginseng using experimentally validated and machine learning-based prediction results. The targets of the network were analyzed in terms of related biological process, pathways, and diseases. The majority of targets were found to be related with primary metabolic process, signal transduction, nitrogen compound metabolic process, blood circulation, immune system process, cell-cell signaling, biosynthetic process, and neurological system process. In pathway enrichment analysis of targets, mainly the terms related with neural activity showed significant enrichment and formed a cluster. Finally, relative degrees analysis for the target-disease association of P. ginseng revealed several categories of related diseases, including respiratory, psychiatric, and cardiovascular diseases.

Postnatal treadmill exercise attenuates prenatal stress-induced apoptosis through enhancing serotonin expression in aged-offspring rats.

Maternal stress during pregnancy affects negative impact on health of offspring. In the present study, we compared the effects of maternal treadmill exercise and offspring treadmill exercise on prenatal stress-induced apoptosis and serotonin expression in offspring. Stress to the pregnant rats was induced by exposure of maternal rats to the hunting dog in an enclosed room. Exposure time was 10 min, three times per day, with a 1-h interval between exposures. This regimen was maintained from the seventh day of gestation until delivery. The pregnant rats in the exercise group were forced to run on a motorized tread-mill for 30 min once a day, started 7 days after pregnancy until delivery. The offspring in the exercise group were forced to run on a motorized treadmill for 30 min once a day, started 4 weeks after birth for 4 weeks. In the present results, offspring exposed to prenatal stress exhibited lower Bcl-2 level and higher Bax level in the hippocampus, lower 5-hydroxytryptamine (5-HT) and tryptophan hydroxylase (TPH) expression in the dorsal raphe, and higher c-Fos expression in the locus coeruleus compared to age-matched control rats. Treadmill exercise of offspring suppressed Bax expression and enhanced Bcl-2 expression in the hippocampus, increased 5-HT and TPH expression in the dorsal raphe, and enhanced c-Fos expression in the locus coeruleus of offspring. Tread-mill exercise of offspring suppressed prenatal stress-induced apoptosis and normalized prenatal stress-induced alterations in serotonin synthesis and neuronal activation. However maternal treadmill exercise during pregnancy exerted no significant effect on offspring.

Effects of treadmill exercise-intensity on short-term memory in the rats born of the lipopolysaccharide-exposed maternal rats.

Maternal infection is an important factor causing neonatal brain injury and later developmental disability. In the present study, we investigated the effects of treadmill exercise intensity on short-term memory, hippocampal neurogenesis, and expression of brain-derived neurotrophic factor (BDNF), and tyrosine kinase receptor B (TrkB) in the rats born of lipopolysaccharide (LPS)-exposed maternal rats. The rats were divided into six groups: control group, mild-intensity exercise group, moderate-intensity exercise group, maternal LPS-exposed group, maternal LPS-exposed and mild-intensity exercise group, maternal LPS-exposed and moderate-intensity exercise group. The rats in the exercise groups were forced to run on a treadmill for 30 min 5 times a week for 4 weeks. The exercise load consisted of running at the speed of 8 m/min for the mild-intensity exercise groups and 14 m/min for moderate-intensity exercise groups. The latency in the step-down avoidance task was deter-mined for the short-term memory. Immunohistochemistry for 5-bro-mo-2'-deoxyuridine was performed to determine hippocampal cell proliferation and neurogenesis. Western blot analysis was performed for the detection of BDNF and TrkB expression. In the present study, tread-mill exercise improved short-term memory deteriorated by maternal LPS exposure. Treadmill exercise increased cell proliferation and neurogenesis in the hippocampal dentate gyrus of the rats born of the LPS-exposed maternal rats. Treadmill exercise increased BDNF and TrkB expression in the hippocampus of the rats born of the LPS-exposed maternal rats. These effects of treadmill exercise were similarly appeared at both mild-intensity and moderate-intensity.

Impact of Several Types of Stresses on Short-term Memory and Apoptosis in the Hippocampus of Rats.

PURPOSE: Stress has a deteriorating effect on hippocampal function. It also contributes to symptom exacerbation in many disease states, including overactive bladder and interstitial cystitis/bladder pain syndrome. We investigated the effects of various types of stresses (restraint, noise, and cold) on short-term memory and apoptosis in relation with corticotropin-releasing factor (CRF) expression. METHODS: Rats in the restraint stress group were restrained in a transparent Plexiglas cylinder for 60 minutes twice daily. Rats in the noise stress group were exposed to the 120 dB supersonic machine sound for 60 minutes twice daily. Rats in the cold stress group were placed in a cold chamber at 4℃ for 60 minutes twice daily. Each stress was applied for 10 days. A step-down avoidance test for short-term memory, immunohistochemistry for caspase-3 expression, and western blot analysis for Bax and Bcl-2 expressions were conducted. RESULTS: Latency time was decreased and CRF expression in the hippocampal dentate gyrus and hypothalamic paraventricular nucleus were increased in all of the stress groups. The number of caspase-3-positive cells in the hippocampal dentate gyrus was increased and the expressions of Bax and Bcl2 in the hippocampus were decreased in all of the stress groups. CONCLUSIONS: All of the stress groups experienced short-term memory impairment induced by apoptosis in the hippocampus. The present results suggest the possibility that these stresses affecting the impairment of short-term memory may also induce functional lower urinary tract disorders.

Effect of anti-siglec-f antibody and reactive oxygen species blocking on histamine release in urinary bladder of ovalbumin-treated mice.

PURPOSE: Sialic acid-binding Ig-like lectin (Siglec) is an immune inhibitory receptor that plays a role in the negative regulation of the activation of immune cells. This study aimed to evaluate the effects of anti-Siglec-F on plasma and urinary histamine levels in ovalbumin (OVA)-challenged urinary bladder in mice. METHODS: Thirty BALB/c mice were used. In group I (control group, n=5), mice were sensitized with OVA and challenged with saline. In group II (OVA challenge group, n=5), OVA was used for intraperitoneal sensitization and intravesical challenge. The challenged mice in group III (control immunoglobulin G [IgG] group, n=5) and those in group IV (anti-Siglec-F group, n=5) were intraperitoneally pretreated with rabbit control IgG or anti-Siglec-F antibody, respectively. In groups V (N-acetylcysteine [NAC] in OVA challenge group, n=5) and VI (control NAC only, n=5), mice were pretreated with NAC. RESULTS: Urinary histamine concentrations were significantly higher 7 days after intravesical OVA challenge (P<0.01), whereas plasma histamine levels were not. Pretreatment with anti-Siglec-F antibody significantly prevented the increase in urinary histamine release (P<0.05), whereas pretreatment with the IgG antibody control did not. Also, pretreatment of the OVA challenge group with NAC did not affect the histamine concentration in either urine or plasma. CONCLUSIONS: Systemic anti-Siglec-F treatment showed anti-allergic effects at least on local histamine release, particularly in the lower urinary bladder.

Protective effect of Acanthopanax senticosus against ethanol-induced apoptosis of human neuroblastoma cell line SK-N-MC.

The protective effect of Acanthopanax senticosus (AS) against ethanol (EtOH)-induced apoptosis of the human neuroblastoma cell line SK-N-MC was investigated via 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay, flow cytometric analysis, DNA fragmentation assay, reverse transcription-polymerase chain reaction (RT-PCR), and caspase-3 assay. It was shown that cells treated with EtOH exhibit classical apoptotic features, while cells pre-treated with Acanthopanax senticosus prior to EtOH exposure showed decreased occurrence of apoptotic features. In addition, Acanthopanax senticosus pre-treatment was shown to inhibit EtOH-induced increase in caspase-3 mRNA expression and activity. These results suggest that Acanthopanax senticosus may exert a protective effect against EtOH-induced apoptosis of human neuroblastoma cells.

Folium mori increases cell proliferation and neuropeptide Y expression in dentate gyrus of streptozotocin-induced diabetic rats.

The possibility has been raised that Folium mori is clinically effective for the treatment and prevention of diabetes. In the present study, the effects of Folium mori on cell proliferation and expression of neuropeptide Y (NPY) in the dentate gyrus of rats with streptozotocin (STZ)-induced diabetes were investigated by 5-bromo-2'-deoxyuridine (BrdU) immunohistochemistry and NPY immunohistochemistry. In rats with STZ-induced diabetes, cell proliferation and NPY expression in the dentate gyrus were suppressed, and treatment with Folium mori was shown to increase new cell formation and NPY expression in the dentate gyrus in both normal rats and those with STZ-induced diabetes. In light of previous studies, this result appears to indicate that increased expression of NPY in the dentate gyrus induced by treatment with Folium mori is associated with the observed effect of Folium mori extract on cell proliferation. Based on the present results, it is suggested that Folium mori treatment may aid in the recovery from the central nervous system complications of diabetes mellitus.

Acupuncture decreases nitric oxide synthase expression in periaqueductal gray area of rats with streptozotocin-induced diabetes.

Acupuncture has been used as a clinical treatment in Oriental medicine for various diseases including diabetes mellitus, one of the most common metabolic disorders in humans. In the present study, the effect of acupuncture on the expressions of neuronal nitric oxide synthase (nNOS) and nitric oxide synthase (NOS) in the dorsolateral periaqueductal gray (DL-PAG) area of rats with streptozotocin (STZ)-induced diabetes was investigated via nNOS immunohistochemistry and nicotinamide adenine dinucleotide phosphate-diaphorase histochemistry. Enhanced expression of nNOS and NOS was detected in the DL-PAG of rats with STZ-induced diabetes, and acupunctural treatment at Zusanli acupoint suppressed the diabetes-induced enhancement in the expression of nNOS and NOS. The present results demonstrate that acupuncture is effective in the modulation of the expression of nNOS and NOS in the DL-PAG under diabetic conditions.