Drug-target interaction (DTI) is fundamental to novel drug research and development (R &D), playing a critical role in elucidating drug mechanisms of action and improving the cost-effectiveness of drug discovery. Although China has seen rapid accumulation of drug-target resources in recent years, the industry still faces challenges such as a shortage of original targets and excessively high target concentration. The rapid advancement of artificial intelligence (AI) has provided an efficient technological pathway for DTI prediction, establishing it as a core tool for accelerating drug discovery. This paper systematically reviews the fundamental data support framework for DTI prediction tasks, elaborating in detail on three core data types— drug characterization, target characterization, and drug-target associations-as well as the classification and application boundaries of mainstream benchmark datasets tailored to different tasks within the field. On this basis, it comprehensively surveys the technological evolution of AI-driven DTI prediction, summarizing the core paradigms and technical advances of traditional machine learning and deep learning methods in a single-modality setting, alongside cutting-edge multimodal approaches that integrate two modalities (cross-subject/within-subject fusion) and three or more modalities for multidimensional information integration. Finally, this paper analyzes the current major challenges in data quality, model generalizability and interpretability, and real-world deployment, while also discussing future trends in standardized dataset construction, generative AI, and large-scale biomedical foundation models, aiming to provide a reference for both research and industrial applications in the AI-driven DTI prediction field.

Objective:To investigate the association between short-term blood pressure variability(BPV)and all-cause mortality in elderly patients with early sepsis-associated acute kidney injury(SA-AKI).Methods:A retrospective study was conducted on elderly patients with early SA-AKI from the Medical Information Mart for Intensive Care Ⅳ(MIMIC-Ⅳ)database between 2008 and 2022.All patients were divided into four groups(Q1-Q4)according to the quartiles of the 24-hour systolic blood pressure coefficient of variation(SBPCV): Q1(SBPCV<8.69%), Q2(8.69%≤SBPCV<10.84%), Q3(10.84%≤SBPCV<13.64%), Q4(SBPCV≥13.64%). Cox regression analysis was used to assess the association between SBPCV and one-year all-cause mortality.Restricted cubic spline regression was used for non-linear testing, and threshold effect analysis was further performed for non-linear relationships.Results:A total of 5 955 elderly patients were included in the study, and 2 746(46.11%)patients died within one year following the diagnosis of early SA-AKI.Patients in groups Q1 to Q4 had one-year all-cause mortality rates of 46.34%(690/1 489), 45.56%(678/1 488), 42.47%(632/1 488), and 50.07%(746/1 490), respectively.After adjusting for relevant confounding variables, Cox regression analysis showed that the Q1 group and Q4 group had a 26%( HR=1.26, 95% CI: 1.13-1.41, P<0.001)and 12%( HR=1.12, 95% CI: 1.01-1.24, P=0.047)higher risk of one-year mortality compared to the Q3 group, respectively.SBPCV and one-year all-cause mortality showed a U-shaped non-linear manner( P for nonlinear<0.001). Threshold effect analysis indicated that when SBPCV<13.07%, each standard deviation increase in SBPCV was associated with a 15% reduction in mortality risk( HR=0.85, 95% CI: 0.78-0.92, P<0.001); conversely, when SBPCV>13.07%, each standard deviation increase in SBPCV was associated with an 11% increase in mortality risk( HR=1.11, 95% CI: 1.03-1.21, P=0.011). Conclusions:Short-term BPV may be associated with long-term all-cause mortality in elderly patients with early SA-AKI in a U-shaped manner.However, these findings require further confirmation through high-quality prospective studies.

Objective:To investigate the association between short-term blood pressure variability(BPV)and all-cause mortality in elderly patients with early sepsis-associated acute kidney injury(SA-AKI).Methods:A retrospective study was conducted on elderly patients with early SA-AKI from the Medical Information Mart for Intensive Care Ⅳ(MIMIC-Ⅳ)database between 2008 and 2022.All patients were divided into four groups(Q1-Q4)according to the quartiles of the 24-hour systolic blood pressure coefficient of variation(SBPCV): Q1(SBPCV<8.69%), Q2(8.69%≤SBPCV<10.84%), Q3(10.84%≤SBPCV<13.64%), Q4(SBPCV≥13.64%). Cox regression analysis was used to assess the association between SBPCV and one-year all-cause mortality.Restricted cubic spline regression was used for non-linear testing, and threshold effect analysis was further performed for non-linear relationships.Results:A total of 5 955 elderly patients were included in the study, and 2 746(46.11%)patients died within one year following the diagnosis of early SA-AKI.Patients in groups Q1 to Q4 had one-year all-cause mortality rates of 46.34%(690/1 489), 45.56%(678/1 488), 42.47%(632/1 488), and 50.07%(746/1 490), respectively.After adjusting for relevant confounding variables, Cox regression analysis showed that the Q1 group and Q4 group had a 26%( HR=1.26, 95% CI: 1.13-1.41, P<0.001)and 12%( HR=1.12, 95% CI: 1.01-1.24, P=0.047)higher risk of one-year mortality compared to the Q3 group, respectively.SBPCV and one-year all-cause mortality showed a U-shaped non-linear manner( P for nonlinear<0.001). Threshold effect analysis indicated that when SBPCV<13.07%, each standard deviation increase in SBPCV was associated with a 15% reduction in mortality risk( HR=0.85, 95% CI: 0.78-0.92, P<0.001); conversely, when SBPCV>13.07%, each standard deviation increase in SBPCV was associated with an 11% increase in mortality risk( HR=1.11, 95% CI: 1.03-1.21, P=0.011). Conclusions:Short-term BPV may be associated with long-term all-cause mortality in elderly patients with early SA-AKI in a U-shaped manner.However, these findings require further confirmation through high-quality prospective studies.

Oxalicine B ( 1) is an α-pyrone meroterpenoid with a unique bispirocyclic ring system derived from Penicillium oxalicum. The biosynthetic pathway of 15-deoxyoxalicine B ( 4) was preliminarily reported in Penicillium canescens, however, the genetic base and biochemical characterization of tailoring reactions for oxalicine B ( 1) has remained enigmatic. In this study, we characterized three oxygenases from the metabolic pathway of oxalicine B ( 1), including a cytochrome P450 hydroxylase OxaL, a hydroxylating Fe(II)/α-KG-dependent dioxygenase OxaK, and a multifunctional cytochrome P450 OxaB. Intriguingly, OxaK can catalyze various multicyclic intermediates or shunt products of oxalicines with impressive substrate promiscuity. OxaB was further proven via biochemical assays to have the ability to convert 15-hydroxdecaturin A ( 3) to 1 with a spiro-lactone core skeleton through oxidative rearrangement. We also solved the mystery of OxaL that controls C-15 hydroxylation. Chemical investigation of the wild-type strain and deletants enabled us to identify 10 metabolites including three new compounds, and the isolated compounds displayed potent anti-influenza A virus bioactivities exhibiting IC₅₀ values in the range of 4.0-19.9 μmol/L. Our studies have allowed us to propose a late-stage biosynthetic pathway for oxalicine B ( 1) and create downstream derivatizations of oxalicines by employing enzymatic strategies.

Objective:To investigate the consistency between the iodine-unstained area and the pathological size of endoscopic submucosal dissection (ESD) specimens of superficial esophageal cancer.Methods:A retrospective study was performed on data of 32 patients with superficial esophageal cancer who accepted ESD from May 2019 to April 2020 in the First Affiliated Hospital, Zhejiang University School of Medicine. The maximum transverse diameter and maximum longitudinal diameter of the iodine-unstained area were compared with the tumor pathological area. A size difference no more than 0.5 cm was considered as conformity, any difference between 0.5 and 1.0 cm was considered as non-conformity, and any difference no less than 1.0 cm was considered as serious non-conformity. At the same time, pink sign after spraying Lugo solution and the consistency of pink sign area with the iodine free area were observed.Results:A total of 32 patients with 33 lesions were enrolled in this study, including 23 males and 9 females and the age of the patients was 59.5±7.3 years. There were 19 (57.6%) lesions whose size of iodine-unstained area was consistent with the tumor pathological area. These 19 lesions were all positive for the pink sign, and the pink sign area overlapped with the iodine-unstained area. In addition, 4 (12.1%) iodine-unstained areas of the lesions did not match the size of the pathological area, and 10 (30.3%) iodine-unstained areas of the lesions were seriously inconsistent with the size of the pathological area. These 14 (42.4%) lesions were all positive for pink sign, and the pink sign area was significantly smaller than the iodine-unstained area. Among the 14 discordant lesions, 2 lesions underwent ESD according to the iodine-unstained area, which resulted in excessive resection and postoperative stenosis.Conclusion:Determining the extent of superficial esophageal cancer by iodine-unstained areas before ESD may lead to excessive resection of the lesions, which is related to the fact that the iodine-unstained areas of the lesions are sometimes significantly larger than the pink sign areas. Therefore, in order to achieve precise treatment, endoscopists can choose the iodine-unstained area with positive pink sign as the first choice for resection.

Objective: To screen and identify serum protein biomarkers for the differential diagnosis between ischemic colitis (IC) and ulcerative colitis (UC) by tandem mass tag (TMT) combined with liquid chromatography/tandem mass spectrometry (LC-MS/MS). Methods: From January 2018 to January 2019, at the First Affiliated Hospital of School of Medicine of Zhejiang University, patients with UC or IC, and health controls, each 10 cases, were enrolled into UC group, IC group and normal control (NC) group, respectively. Fasting serum samples of all the subjects were collected. After removal of high-abundance protein, followed by proteolysis, peptide labeling and fractionating, the samples were then processed by mass spectrometry. The protein with TMT data of three groups was obtained and protein with TMT value 0 were removed. Heat map of protein was constructed. The differential protein was defined as the protein fold change over 1.5 or less than 0.67. The Reactome database was used to cluster the pathways of differential proteins among groups. Statistical methods included t test, hypergeometry test and corrected by BH multiple test. Results: A total of 357 serum proteins were identified by proteomic profiling. There were 27 differential proteins between the IC group and the NC group, including six up-regulated proteins and 21 down-regulated proteins. There were 228 differential proteins between the UC group and the NC group, including 75 up-regulated proteins and 153 down-regulated proteins. There were 49 differential proteins between UC group and IC group, including 22 up-regulated proteins and 27 down-regulated proteins. In the comparison of differential proteins between the NC group, IC group and UC group, only the expression of fibrin 3 was statistically significant (the fold change between UC and NC, between UC and IC, between IC and NC were 0.24, 0.46 and 0.53, respectively; t=-5.089, -7.298 and -3.919, all P<0.01). The results of pathway cluster analysis showed that in the comparison of differential proteins between IC group and NC group, only the platelet degranulation pathway was enriched, and 10 proteins were involved in this pathway (P<0.01). In the comparison of differential proteins between UC group and NC group, there were 58 pathways enriched, of which 38 proteins were involved in the platelet degranulation pathway, 16 proteins were involved in the initial complement trigger pathway, 13 proteins were involved in the complement cascade pathway, and 11 proteins were involved in antibody-mediated complement activation pathway (all P<0.01). In the comparison of differential proteins between UC group and IC group, three different pathways were obtained. Among them, nine proteins were involved in the platelet degranulation pathway, seven proteins were involved in the initial complement trigger pathway, and five proteins were involved in the complement cascade pathway (all P<0.01). Conclusions The difference in serum proteome between IC patients and UC patients was significant, and the differential proteins are mainly involved in platelet activation and complement activation. The candidate proteins identified in this study may be used as biomarkers for the differential diagnosis of UC and IC in the future.

Objective To explore the clinical,endoscopic,pathologic and prognostic characteristics of primary gastric lymphoma (PGL) and to improve the level of diagnosis and treatment.Methods Sixtythree patients who were confirmed as PGL with operation and endoscopic biopsy pathology during January 2001 to December 2010 were retrospectively analyzed with respects of clinical,endoscopic and pathologic features.Survival analysis and prognosis were evaluated by kaplan-Meier and Cox proportional hazard model,respectively.Results In 63 PGL patients,the numbers of male and female were 40 and 23,respectively,and the average age was (59.8±13.3)years.The major symptoms were abdominal pain,abdominal distension,and gastrointestinal hemorrhage,accounting for 47.6 ％ (30/63),17.5 ％ (11/63),and 17.5 ％ (11/63),respectively.There were 39 (61.9 ％) PGL patients with endoscopic performance for ulcers,34 (54.0 ％) cases involved the gastric stomach antrum.The most immunohistochemistry analyses were diffuse large B-cell lymphoma (DLBCL) (71.4 ％,45/63),followed by mucosa-associated lymphoid tissue (MALT) lymphoma (22.2％,14/63).The frequency of Helicobacter pylori (H.pylori) positivity was lower in patients with DLBCL than that in patients with MALT lymphoma (37.8％(17/45) vs 10/14,x2 =4.872,P=0.027).The accumulate survival rates of one,three and five years were 74.6％,63.5％,55.6％,respectively,and the average survival time was (41.5±3.0) months (95％ confidence interval (CI) 35.7 to 47.4 months) in PGL patients.There was no difference in the average survival time between DLBCL patients treated with surgery combined chemotherapy and those with surgery or chemotherapy alone (38.33±5.21) months vs (50.17±8.98) months vs (41.39±4.40) months,P＞0.05).The patients diagnosed as DLBCL with H.pylori positive had longer average survival time than those with H.pylori negative ((51.90±4.30) months vs (33.30±4.50) months,t=-4.004,P＜0.01).Conclusions Male patients with PGL are slightly more than female.Abdominal pain is the most frequent symptom.Ulcerative lesions are the most common endoscopic demonstrations mostly at stomach sinus.DLBCL is the most pathologic characteristic.There is no significant difference in the survival rate between patients treated with surgery combined with chemotherapy and those treated with surgery or chemotherapy alone.

Objective To analyze the clinical manifestations, computed tomography scan (CT), gastroscope, endo-scopic ultrasonography (EUS), and therapy method of gastritis cystica profunda. Methods Retrospectively analyzed clinical manifestations, CT, gastroscope, EUS, and pathological results of 6 cases of gastritis cystica profunda. Results In these 6 cases, 3 of them were doubted gastric carcinoma, 3 cases were considered stomach mass by CT. Gastroscope hinted apophysis lesions, but all cases were suggested gastritis cystica profunda by EUS. And all cases were removed through endoscopic submucosal dissection (ESD). Pathology were confirmed the diagnosis. Conclusion EUS combined with endoscopic mucosal resection (EMR) or ESD technique can improve the diagnostic rate. For gas-tritis cystica profunda which are not associated with malignant tumor can be treated through ESD.

To establish a cell-based rapid luciferase suppression assay for high-throughput screening (HTS) anti-alphaviruses compounds screening, which could cause viral encephalitis, raise the social issues associated directly with public health and huge economic burden to the society. The Gaussia luciferase assay system was used for HTS model for identifying inhibitors of labeled virus XJ160-GLUC. The decreased 50% GLUC activity inhibition ratio was deemed to be the screening positive index. The reaction system in this model was optimized, and the reliability of the model was evaluated. For HTS model's optimization, cells were infected with XJ160-GLUC at an MOI of 0.025 PFU/cell. The supernatant treated with compounds 48h were collected for GLUC expression detection. In the model, Z' factor was up to 0.71, demonstrating that HTS assay for identifying inhibitors that target all aspects of the viral life cycle of XJ160-GLUC was stable and reliable. After screening 8080 compounds (five-in-one), 341 positive samples were selected, and the positive rate was 4.2% with a cutoff at 50% inhibition. Then 1705 compounds were screened subsequently and the positive rate was 1.1% with obtaining 19 positive compounds. These results will lay the foundation for finding the anti-alphaviruses' drug targets.
Alphavirus ; drug effects ; genetics ; metabolism ; Animals ; Antiviral Agents ; pharmacology ; Drug Evaluation, Preclinical ; methods ; Genes, Reporter ; High-Throughput Screening Assays ; methods ; Luciferases ; genetics ; metabolism

Objective To explore the diagnostic model and clinical application value of serum proteomic fingerprint in inflammatory bowel disease (IBD) .Methods Serum proteome profiles of 72 IBD patients (54 Crohn′s disease (CD) and 18 ulcerative colitis (UC) and 44 healthy controls were analyzed by the weak cation exchange (WCX) beads combined matrix‐assisted laser desorption/ionization time of flight mass spectrometry (MALDI‐TOF‐MS ) technique . Among three groups , every two groups were compared .Wilcoxon rank sum test was used to screen out the peaks of difference expressed protein (P<0 .05) .Genetic algorithm combining with support vector machine (SVM ) was utilized to select the best diagnostic model .The predictive effects of this model was evaluated by leave one out method (LOO ) . Results The 10 most discriminating protein peaks were screened out between CD group and healthy control group , between UC group and healthy control group , between CD group and UC group . A diagnostic model established with four protein peaks ,the mass‐to‐charge ratio (M /Z ) of them was 3 275 .29 ,4 963 .91 ,4 980 .53 and 5 336 .90 ,could better distinguish CD and healthy controls .The specificity was 97 .7% ,and the sensitivity was 92 .6% in CD diagnosis .A diagnostic model established with four protein peaks ,the M /Z of them was 2 272 .41 ,2 660 .42 ,3 029 .77 and 5 002 .78 ,could better distinguish UC and healthy controls .The specificity was 100 .0% ,and the sensitivity was 94 .4% .A specificity was 50 .0% and sensitivity was 88 .9% in CD diagnosis with the diagnostic model of six protein peaks and the M /Z of them was 2 082 .63 ,2 210 .64 ,4 039 .02 ,4 298 .30 ,4 978 .03 ,5 002 .22 .Conclusion The diagnostic model of serum difference expressed protein in CD and UC is established by MALDI‐TOF‐MS technique and genetic algorithm combining with SVM ,which has high diagnostic value in IBD .