Transformer models have emerged as pivotal tools within the realm of drug discovery, distinguished by their unique architectural features and exceptional performance in managing intricate data landscapes. Leveraging the innate capabilities of transformer architectures to comprehend intricate hierarchical dependencies inherent in sequential data, these models showcase remarkable efficacy across various tasks, including new drug design and drug target identification. The adaptability of pre-trained transformer-based models renders them indispensable assets for driving data-centric advancements in drug discovery, chemistry, and biology, furnishing a robust framework that expedites innovation and discovery within these domains. Beyond their technical prowess, the success of transformer-based models in drug discovery, chemistry, and biology extends to their interdisciplinary potential, seamlessly combining biological, physical, chemical, and pharmacological insights to bridge gaps across diverse disciplines. This integrative approach not only enhances the depth and breadth of research endeavors but also fosters synergistic collaborations and exchange of ideas among disparate fields. In our review, we elucidate the myriad applications of transformers in drug discovery, as well as chemistry and biology, spanning from protein design and protein engineering, to molecular dynamics (MD), drug target identification, transformer-enabled drug virtual screening (VS), drug lead optimization, drug addiction, small data set challenges, chemical and biological image analysis, chemical language understanding, and single cell data. Finally, we conclude the survey by deliberating on promising trends in transformer models within the context of drug discovery and other sciences.

OBJECTIVE: This study aimed to evaluate the association between susceptibility genes and non-alcoholic fatty liver disease (NAFLD) in children with obesity. METHODS: We conducted a two-step case-control study. Ninety-three participants were subjected to whole-exome sequencing (exploratory set). Differential genes identified in the small sample were validated in 1,022 participants using multiplex polymerase chain reaction and high-throughput sequencing (validation set). RESULTS: In the exploratory set, 14 genes from the NAFLD-associated pathways were identified. In the validation set, after adjusting for sex, age, and body mass index, ECI2 rs2326408 (dominant model: OR = 1.33, 95% CI: 1.02-1.72; additive model: OR = 1.22, 95% CI: 1.01-1.47), C6orf201 rs659305 (dominant model: OR = 1.30, 95% CI: 1.01-1.69; additive model: OR = 1.21, 95% CI: 1.00-1.45), CALML5 rs10904516 (pre-ad dominant model: OR = 1.36, 95% CI: 1.01-1.83; adjusted dominant model: OR = 1.40, 95% CI: 1.03-1.91; and pre-ad additive model: OR = 1.26, 95% CI: 1.04-1.66) polymorphisms were significantly associated with NAFLD in children with obesity ( P < 0.05). Interaction analysis revealed that the gene-gene interaction model of CALML5 rs10904516, COX11 rs17209882, and SCD5 rs3733228 was optional ( P < 0.05), demonstrating a negative interaction between the three genes. CONCLUSION In the Chinese population, the CALML5 rs10904516, C6orf201 rs659305, and ECI2 rs2326408 variants could be genetic markers for NAFLD susceptibility.
Humans ; Non-alcoholic Fatty Liver Disease/genetics* ; Child ; Male ; Female ; Genetic Predisposition to Disease ; Case-Control Studies ; Exome Sequencing ; Adolescent ; Polymorphism, Single Nucleotide ; Obesity/complications* ; Pediatric Obesity/complications* ; China

Transformer models have emerged as pivotal tools within the realm of drug discovery,distinguished by their unique architectural features and exceptional performance in managing intricate data landscapes.Leveraging the innate capabilities of transformer architectures to comprehend intricate hierarchical dependencies inherent in sequential data,these models showcase remarkable efficacy across various tasks,including new drug design and drug target identification.The adaptability of pre-trained trans-former-based models renders them indispensable assets for driving data-centric advancements in drug discovery,chemistry,and biology,furnishing a robust framework that expedites innovation and dis-covery within these domains.Beyond their technical prowess,the success of transformer-based models in drug discovery,chemistry,and biology extends to their interdisciplinary potential,seamlessly combining biological,physical,chemical,and pharmacological insights to bridge gaps across diverse disciplines.This integrative approach not only enhances the depth and breadth of research endeavors but also fosters synergistic collaborations and exchange of ideas among disparate fields.In our review,we elucidate the myriad applications of transformers in drug discovery,as well as chemistry and biology,spanning from protein design and protein engineering,to molecular dynamics(MD),drug target iden-tification,transformer-enabled drug virtual screening(VS),drug lead optimization,drug addiction,small data set challenges,chemical and biological image analysis,chemical language understanding,and single cell data.Finally,we conclude the survey by deliberating on promising trends in transformer models within the context of drug discovery and other sciences.

Objective To investigate the protective effects of Qiji Special Dietary Food on exercise-induced myocardial injury in mice.Methods An exercise-induced myocardial injury model was established using a treadmill running protocol,and at the same time,the modeled mice were administered Qiji Special Dietary Food via oral gavage,followed by a 4-week treadmill exhaustion test.Serum levels of cardiac troponin T(cTnT),creatine kinase(CK),and lactate dehydrogenase(LDH)were measured.Myocardial tissues were analyzed for superoxide dismutase(SOD)activity and malondialdehyde(MDA)content.Histopathological alterations and ultrastructural changes in myocardial tissue were evaluated using light microscopy and transmission electron microscopy(TEM).Results After 4 weeks of exhaustive training,compared to the control group,the model group exhibited significantly elevated serum cTnT,CK,and myocardial MDA levels(P<0.01),along with reduced myocardial SOD activity(P<0.01).Compared to the model group,high-,medium-,and low-dose treatment significantly attenuated the exhaustive exercise-induced increases in serum cTnT level and myocardial MDA content(P<0.01),restoring these indicators to the levels comparable to those of the normal control group(P<0.05).Additionally,all treatment groups had markedly increased myocardial SOD activity,with no significant difference from the normal group.Histopathological and ultrastructural analyses revealed markedly alleviated myocardial damage in the treatment groups,with the medium-dose group exhibiting the most pronounced protective effects.Conclusion Qiji Special Dietary Food demonstrates significant protective effects against exercise-induced myocardial injury in mice,which maybe associated with its antioxidant activity and mitigation of oxidative stress.

Objective To systematically evaluate the effectiveness and acceptability of antidepressant drugs in the treatment of postpartum depression(PPD).Methods The PubMed,Cochrane Library,Embase,Web of Science,China National Knowledge Infrastructure(CNKI),Wanfang Database,VIP Journals of Chinese Scienc were searched,and Chinese Biomedical Literature Service System(SinoMed)database until November 2023.Screen randomized controlled trials(RCTs)of antidepressant drugs for the treatment of PPD.The treatment group was given antidepressant drugs,and the control group was given placebo or another antidepressant drug.Meta-analysis of effectiveness and acceptability is performed using Stata 17.0 software.Results A total of 27 RCTs with a total of 2 202 patients were included.The results of meta-analysis showed:The top three efficacy relative to placebo were mirtazapine[odds ratio(OR)=2.25,95％confidence interval(CI)=(1.20-3.30),P＜0.05],nortriptyline[OR=1.50,95％CI=(0.55-2.44),P＞0.05],venlafaxine[OR=1.35,95％CI=(0.13-2.56),P＞0.05].Acceptability is compared with placebo in the top three Chinese herbal medicine[OR=0.47,95％CI=(-0.72-1.66),P＞0.05],nortriptyline[OR=-0.08,95％CI=(-1.16-1.33),P＞0.05],venlafaxine[OR=-0.12,95％CI=(-1.47-1.24),P＞0.05].Conclusion Nortriptyline,venlafaxine,trazodone,and duloxetine are effective in treating PPD without obvious adverse drug reactions.

Objective To propose a lead attention network-based ST-T change recognition algorithm to detect ECG ST-T changes accurately.Methods Firstly,heartbeat signals were extracted through R-wave localization,and a 12-lead heartbeat matrix was generated by correlation-based screening and merging to realize data augmentation.Secondly,a lead attention module was constructed by combining depthwise convolution(DWConv)with the channel attention squeeze-and-excitation block(SE-block)structure to perceive the differences in ST-T status among electrocardiogram leads.Thirdly,the mapping output by two independent attention modules was fused and splicing with the original signal residual was carried out,so that attention information extraction and original information transfer were enhanced effectively.Finally,SE-ResNet was used as the backbone network to extract signal features to complete the classification and identification of ST-T changes.To validate the recognition performance of the proposed algorithm for ST-T changes in ECG,the 12-lead ECG data of 97 472 patients containing different ECG rhythms were collected for ablation and comparison experiments at the First Affiliated Hospital of Army Medical University.Results The proposed algorithm achieved an AUC of 0.965 with a sensitivity of 90.51%,specificity of 90.23%,positive predictive value of 89.24%and overall accuracy of 90.36%on an independent test set.Comparative analysis demonstrated superior performance to four benchmark architectures,including VGG16,ResNet18,MobileNetV3-Small and ShuffleNet,in terms of both classification accuracy and computational efficiency.Conclusion The algorithm designed can accurately detect ST-T changes and can be used for wearable ECG automatic analysis to assist in the early warning of cardiovascular diseases in both acute and chronic patients and highland residents.[Chinese Medical Equipment Journal,2025,46(7):1-11]

Objective To systematically evaluate the effectiveness and acceptability of antidepressant drugs in the treatment of postpartum depression(PPD).Methods The PubMed,Cochrane Library,Embase,Web of Science,China National Knowledge Infrastructure(CNKI),Wanfang Database,VIP Journals of Chinese Scienc were searched,and Chinese Biomedical Literature Service System(SinoMed)database until November 2023.Screen randomized controlled trials(RCTs)of antidepressant drugs for the treatment of PPD.The treatment group was given antidepressant drugs,and the control group was given placebo or another antidepressant drug.Meta-analysis of effectiveness and acceptability is performed using Stata 17.0 software.Results A total of 27 RCTs with a total of 2 202 patients were included.The results of meta-analysis showed:The top three efficacy relative to placebo were mirtazapine[odds ratio(OR)=2.25,95％confidence interval(CI)=(1.20-3.30),P＜0.05],nortriptyline[OR=1.50,95％CI=(0.55-2.44),P＞0.05],venlafaxine[OR=1.35,95％CI=(0.13-2.56),P＞0.05].Acceptability is compared with placebo in the top three Chinese herbal medicine[OR=0.47,95％CI=(-0.72-1.66),P＞0.05],nortriptyline[OR=-0.08,95％CI=(-1.16-1.33),P＞0.05],venlafaxine[OR=-0.12,95％CI=(-1.47-1.24),P＞0.05].Conclusion Nortriptyline,venlafaxine,trazodone,and duloxetine are effective in treating PPD without obvious adverse drug reactions.

Objective To propose a lead attention network-based ST-T change recognition algorithm to detect ECG ST-T changes accurately.Methods Firstly,heartbeat signals were extracted through R-wave localization,and a 12-lead heartbeat matrix was generated by correlation-based screening and merging to realize data augmentation.Secondly,a lead attention module was constructed by combining depthwise convolution(DWConv)with the channel attention squeeze-and-excitation block(SE-block)structure to perceive the differences in ST-T status among electrocardiogram leads.Thirdly,the mapping output by two independent attention modules was fused and splicing with the original signal residual was carried out,so that attention information extraction and original information transfer were enhanced effectively.Finally,SE-ResNet was used as the backbone network to extract signal features to complete the classification and identification of ST-T changes.To validate the recognition performance of the proposed algorithm for ST-T changes in ECG,the 12-lead ECG data of 97 472 patients containing different ECG rhythms were collected for ablation and comparison experiments at the First Affiliated Hospital of Army Medical University.Results The proposed algorithm achieved an AUC of 0.965 with a sensitivity of 90.51%,specificity of 90.23%,positive predictive value of 89.24%and overall accuracy of 90.36%on an independent test set.Comparative analysis demonstrated superior performance to four benchmark architectures,including VGG16,ResNet18,MobileNetV3-Small and ShuffleNet,in terms of both classification accuracy and computational efficiency.Conclusion The algorithm designed can accurately detect ST-T changes and can be used for wearable ECG automatic analysis to assist in the early warning of cardiovascular diseases in both acute and chronic patients and highland residents.[Chinese Medical Equipment Journal,2025,46(7):1-11]

Objective:To analyze clinical, immunopathological, therapeutic, and prognostic features of epidermolysis bullosa acquisita (EBA) .Methods:A retrospective study was conducted on patients with confirmed EBA at the Department of Dermatology, West China Hospital, Sichuan University from January 1, 2015 to July 30, 2022. Their clinical, immunopathological, therapeutic and prognostic features were analyzed. The autoimmune bullous skin disorder intensity score (ABSIS) was used to assess the severity of lesions in patients with EBA, and the visual analogue scale (VAS) to assess itch intensity. Descriptive statistical analysis was primarily carried out, and the correlation between disease severity scores and itch scores was analyzed using Pearson correlation analysis.Results:A total of 13 patients with EBA were included, including 9 males and 4 females, with the age at the clinic visit being 49.0 ± 20.6 years and ABSIS scores being 24.2 ± 10.7 points. One patient was diagnosed with classical EBA, while the remaining 12 patients with inflammatory EBA. Mucosal involvement was observed in 6 cases, whose oral mucosae were all affected. All patients had itching to varying degrees, with VAS scores of 5.6 ± 2.2 points; 9 of the 12 inflammatory EBA patients had VAS scores of ≥ 5 points, whereas 1 classical EBA patient had a VAS score of 2 points; there was no significant correlation between the ABSIS scores and VAS scores ( r = -0.02, P > 0.05). Histopathological examination showed subepidermal cleavages or blister formation and varying degrees of perivascular inflammatory cell infiltration in the superficial dermis of patients with inflammatory EBA. Direct immunofluorescence assay demonstrated linear IgG deposits along the basement membrane zone in all 13 patients, including 12 with concomitant linear C3 deposits in the basement membrane zone, 5 with linear IgA deposits, and 2 with IgM deposits. Indirect immunofluorescence on salt-split skin showed IgG deposition on the dermal side of the salt-split skin in the 13 patients. An elevated eosinophil count in the peripheral blood was observed in 1 out of 11 patients, while increased total IgE levels were noted in 3 out of 9 patients. Among the 13 EBA patients, 11 were treated with systemic glucocorticoids (equivalent to 10 - 100 mg/d of prednisone), and the other 2 were treated with compound glycyrrhizin tablets, sulfasalazine, hydroxychloroquine sulfate, and minocycline hydrochloride alone or in combination. During the follow-up period of 34.0 (27.5, 66.0) months in the 13 patients, 8 achieved complete remission after drug withdrawal, 2 achieved complete remission on therapy, 1 achieved partial remission on minimal therapy, and 2 presented with uncontrolled condition. The time to complete remission off/on therapy was 6.0 (3.8, 17.5) months. Conclusions:The inflammatory phenotype seems to be relatively common in EBA patients, with itching to varying degrees, and oral mucosa was the most commonly involved mucosa in those with mucosal damage. After treatment with systemic glucocorticoids alone or in combination with immunomodulators, most patients could achieve complete remission.

Objective To explore the effects of circ_100284 in affecting the invasion of esophageal squamous cell carcinoma(ESCC)cells and their sensitivity to 5-fluorouracil(5-FU)chemotherapy via regulating miR-217/mitogen-activated protein kinase 1(MAPK1)acting as a competing endogenous RNA(ceRNA).Methods The bioinformatics approach was used to analyze the differentially expressed circRNAs in ESCC.qRT-PCR was used to detect the expression of circ_100284 in ESCC tissues and cells.The 5-FU-resistant KYSE450 cell line(KYSE45-R)was established by increasing concentrations of 5-FU.The IC50 of 5-FU in KYSE450 and KYSE450-R cells was determined through MTT assay.qRT-PCR was used to detect the expression of circ_100284,miR-217,and MAPK1 mRNA in KYSE450 and KYSE450-R cells,while Western blotting detecting the protein expression of MAPK1.In the experiments with KYSE450-R cells,we set up the following groups:(1)blank group(without treatment),si-NC group(transfected with si-NC),si-circ_100284 group(transfected with si-circ_100284),pc-Control group(transfected with pc-Control),pc-circ_100284 group(transfected with pc-circ_100284),pc-circ_100284+mimic NC group(transfected with pc-circ_100284 and mimic NC),and pc-circ_100284+miR-217 mimic group(transfected with pc-circ_100284 and miR-217 mimic).These groups were subjected to MTT assay to detect cell viability,Transwell assay to detect cell invasion,and flow cytometry to detect cell apoptosis.(2)Blank group(without treatment),si-NC group(transfected with si-MAPK1 negative control),si-MAPK1 group(transfected with si-MAPK1),si-MAPK1+inhibitor NC group(transfected with si-MAPK1 and inhibitor NC),and si-MAPK1+miR-217 inhibitor group(transfected with si-MAPK1 and miR-217 inhibitor).We detected the mRNA and protein expression of MAPK1 using qRT-PCR and Western blotting.We evaluated cell viability using MTT assay,invasion with Transwell assay,and apoptosis by flow cytometry.circ_100284-WT or circ_100284-MUT reporter plasmids,as well as MAPK1-WT or MAPK1-MUT reporter plasmids,were co-transfected with miR-NC or miR-217 mimic into KYSE450-R cells for 48 h,and dual luciferase reporter assay was used to measure luciferase activity.Results The bioinformatics analysis revealed significant upregulation of circ_100284 in ESCC.Compared with adjacent normal tissues,the expression of circ_100284 in ESCC tissues is enhanced(P<0.05);compared with the HECC cells,the TE-11,ECA109 and KYSE450 ESCC cell lines showed enhanced expression of circ_100284(P<0.05).Compared with the KYSE450 cells,KYSE450-R cells demonstrated increased IC50 with enhanced expression of circ_100284 and MAPK1 but suppressed expression of miR-217(P<0.05).Compared with the si-NC group,the si-circ_100284 group demonstrated inhibited invasion and proliferation of cells with increased apoptosis(P<0.05).Compared with pc-Control group,the invasion and proliferation of cells in the pc-circ_100284 group are increased,and cell apoptosis is decreased(P<0.05).Over-expression of miR-217 reversed the malignant biological behavior of ESCC cells induced by pc-circ_100284(P<0.05).Compared with si-NC group,in the si-MAPK1 group,we observed decreased cell invasion and proliferation,and increased apoptosis(P<0.05),but miR-217 inhibitor reversed the effect of si-MAPK1 on the biological behavior of ESCC cells(P<0.05).The targeting relationship of circ_100284 and miR-217,miR-217 and MAPK1 is confirmed.Conclusion circ_100284 promotes ESCC cell invasion by regulating miR-217/MAPK1,inhibits the chemosensitivity of ESCC cells to 5-FU,and acts as a tumor-promoting factor in ESCC.