Natural product-based drug combinations (NPDCs) present distinctive advantages in treating complex diseases. While high-throughput screening (HTS) and conventional computational methods have partially accelerated synergistic drug combination discovery, their applications remain constrained by experimental data fragmentation, high costs, and extensive combinatorial space. Recent developments in artificial intelligence (AI), encompassing traditional machine learning and deep learning algorithms, have been extensively applied in NPDC identification. Through the integration of multi-source heterogeneous data and autonomous feature extraction, prediction accuracy has markedly improved, offering a robust technical approach for novel NPDC discovery. This review comprehensively examines recent advances in AI-driven NPDC prediction, presents relevant data resources and algorithmic frameworks, and evaluates current limitations and future prospects. AI methodologies are anticipated to substantially expedite NPDC discovery and inform experimental validation.
Artificial Intelligence ; Biological Products/chemistry* ; Humans ; Drug Combinations ; Drug Discovery/methods* ; Machine Learning ; Algorithms

Objective: To explore the feasibility of clinical application of common day surgery in general surgery. Methods: From January 2018 to December 2018, the clinical data of 40 cases of gallbladder stones(LC group), 40 cases of breast lumps(breast lumps group) and 36 cases of mixed hemorrhoids(mixed hemorrhoids group) admitted to the daytime center of the First People's Hospital of Huainan were retrospectively analyzed.The average hospitalization days and expenses were analyzed, and each group was equally divided into control group(traditional surgery) and research group(day surgery). The hospitalization time and hospitalization expenses were compared between the control group and research group. Results: The hospitalization time of the research group in the LC group, breast lumps group and mixed hemorrhoids group was (21.07±6.31)h, (45.60±7.39)h, (21.07±6.31)h, respectively, which were significantly shorter than those of the control group[(106.80±25.20)h, (184.50±25.00)h, (236.00±54.14)h], the differences were statistically significant(t=15.093, 14.898, -15.645, all P<0.05). The hospitalization expenses of the research group in the LC group, breast lumps group and mixed hemorrhoids group were (2 083.01±436.04)CNY, (8 842.46±861.15)CNY, (3 186.47±464.93)CNY, respectively, which were also significantly lower than those of the control group[(3 594.82±1 059.60)CNY, (11 003.76±1 706.78)CNY, (6 419.69±817.45)CNY], the differences were statistically significant(t=5.245, 6.666, -16.082, all P<0.05). Conclusion Day surgery has obvious advantages and development space in clinical practice of common surgery in general surgery.It has incomparable effect and is superior to traditional surgery.

Objective To investigate the safety and feasibility of the reversed π-shaped esophagojejunal anastomosis in laparoscope-assisted total gastrectomy (LATG) for gastric cancer (GC).Methods The retrospective corss-sectional study was conducted.The clinicopathological data of 18 GC patients who were admitted to the Subei People's Hospital of Jiangsu Province between January 2015 and October 2016 were collected.All the 18 GC patients underwent LATG,surgical procedures included free stomach and lymph node dissection firstly,side-to-side jejunal anastomosis secondly and laparoscopic gastrointestinal reconstruction using reversed π-shaped anastomosis finally.Observation indicators:(1) surgical situations:side-to-side jejunal anastomosis method,conversion to open surgery,operation time,reversed π-shaped anastomosis time,volume of intraoperative blood loss and number of lymph node dissected;(2) postoperative situations:time for initial out-of-bed activity,time to initial anal exsufflation,time for postoperative water intake,time of drainage tube removal,postoperative complications and duration of hospital stay;(3) postoperative pathological examination;(4) follow-up and survival situations.Follow-up using outpatient examination and telephone interview was performed to detect diet intake,anastomosis patency,gastrointestinal obstruction and patients' survival up to March 2017.Measurement data with normal distribution were represented as (x)±s.Results (1) Surgical situations:18 patients underwent successful LATG and reversed π-shaped esophagojejunal anastomosis,without conversion to open surgery and perioperative death.Five patients used in vitro hand-sewn side-to-side esophagojejunal anastomosis through small incision of specimens sampling,and 13 completed all surgery under laparoscopy.Operation time,reversed π-shaped anastomosis time,volume of intraoperative blood loss and number of lymph node dissected of 18 patients were (187±12)minutes,(37±5) minutes,(735±18)mL and 29±2,respectively.(2) Postoperative situations:time for initial out-of-bed activity,time to initial anal exsufflation,time for postoperative water intake and time of drainage tube removal in 18 patients were (1.6±0.5) days,(2.3±0.4) days,(2.5±0.5) days and (7.5± 1.5) days,respectively.One patient complicated with esophagojejunal anastomosis fistula at postoperative day 3 was cured by drainage and symptomatic treatment and then discharged from hospital.Eighteen patients received regularly angiography using oral water-soluble contrast medium after recovering fluid diet intake,showing anastomosis patency and no contrast medium leakage,and then discharged from hospital.Duration of hospital stay of 18 patients was (12± 11) days.(3) Postoperative pathological examination:of 18 patients,15 were diagnosed with adenocarcinoma and 3 with signet-ring cell carcinoma.T2,T3 and T4 of T staging were respectively detected in 3,3 and 12 patients.N0,N1,N2 and N3 of N staging were respectively detected in 8,3,2 and 5 patients.Stage Ⅰ,Ⅱ and Ⅲ of TNM stage were detected in 3,5 and 10 patients,respectively.(4) Follow-up and survival situations:17 of 18 patients were followed up for 6-25 months,with a median time of 12 months.During the follow-up,2 patients were complicated with sour regurgitation and vomiting after eating at month 6 and 12 postoperatively and received gastrointestinal contrast examination,showing anastomotic stenosis,and then were cured by endoscopic dilation and discharged form hospital.Other patients had good diet and survival,without anastomotic complications.Conclusion The reversed π-shaped esophagojejunal anastomosis in LATG for GC is safe and feasible,with good short-term outcomes.

Objective To assess the effect of neutrophil?to?lymphocyte ratio ( NLR) on the recurrence rate of patients with alpha?fetoprotein (AFP) ?negative hepatocellular carcinoma (HCC) and the value of NLR in predicting prognosis. Methods The clinical data of seventy?seven patients diagnosed with AFP?negative HCC and treated with hepatocellular carcinoma surgery in the First Hospital in Weinan from June 2015 to March 2017 were analyzed. According to the recurrence at the end of the follow?up, the patients were divided into the recurrence group and the non?recurrence group. Cox single factor analysis was used to analyze the relationship between the clinicopathological features and postoperative recurrence, and the clinical risk factors with statistically significance in the univariate analysis were placed in the Cox multivariate regression analysis to determine whether it is independent risk factor. Results The differences between the two groups in the number of tumors (31/15,28/3),tumor size (>5 cm) (5. 53±1. 83,4. 65±1. 73),portal vein tumor thrombus (18/28,2/29),microvascular invasion (14/32,2/29) were all statistically significant (P<0. 05). Cox univariate analysis showed that preoperative NLR levels ( RR=1. 125, 95%CI 1. 052-1. 203, P=0. 029 ) , tumor number ( RR=0. 943,95%CI 1. 007-1. 330, P=0. 019 ) , tumor size (>5 cm ) ( RR=0. 550, 95%CI 0. 316-0. 956, P=0. 038),portal vein tumor thrombus (RR=1. 294,95%CI 1. 208-1. 386,P=0. 022),microvascular invasion (RR=1. 575,95%CI1. 209-2. 052,P=0. 028) were the risk factors of postoperative recurrence. Cox regression model showed that tumor number (RR=1. 830,95%CI 1. 184-2. 828,P=0. 026),portal vein tumor thrombus ( RR=2. 860,95%CI 2. 062-3. 968,P=0. 001) ,microvascular invasion ( RR=1. 760,95%CI 1. 019-3. 041,P=0. 037) and preoperative NLR level ( RR=1. 890,95%CI 1. 056-3. 383,P=0. 028) were independent risk factors of the recurrence in AFP negative HCC patients after surgery. Among the 77 patients,46 cases were in the recurrent group, the average value of NLR was 3. 49 ± 0. 30, and the average preoperative NLR of the non?recurrence group ( 31 patients ) were 3. 01 ± 0. 30, the difference between the two groups in NLR value was statistically significant (t=-6. 885,P=0. 000). According to the ROC curve,the NLR=3. 17 corresponded to the maximum Youden index,the sensitivity of NLR was 82. 6%,the specificity was 67. 7%. Conclusion The preoperative NLR level is inversely proportional to the recurrence?free survival time of patients with AFP?negative HCC,which is one of the independent risk factors for recurrence. The optimal critical value of NLR is 3. 17.

Objective To investigate the interference of thyroglobulin antibodies ( TgAb ) on the measurement of thyroglobulin ( Tg) by 2 chemiluminescence immunoassays ( CLIAs) .Methods Data of 199 315 individuals with determined TgAb and Tg , including physical checkup subjects , differentiated thyroid carcinoma ( DTC) patients and patients with other diseases , were retrospectively collected in Peking Union Medical College Hospital from November 2012 to April 2015.The correlation between serum Tg level and serum TgAb concentration was analyzed and the positive rate of TgAb in physical checkup subjects was calculated.Furthermore, 290 serum samples with different TgAb concentration were applied in the recovery test by adding in confirmed serum Tg .The correlation between the recovery of confirmed serum Tg and TgAb concentrations was evaluated using Pearson correlation analysis .Results The serum Tg was all decreased with the elevated TgAb concentration in each group of subjects .The positive rate of TgAb was 10.84%(8 416/77 634) in physical checkup subjects .It was higher in females than in males and was increased with age.Recovery test showed that the average recoveries of confirmed serum Tg in TgAb-negative serum were 107.28%(86.30%-117.60%) and 107.94% (85.60%-124.10%) respectively in Roche and Beckman systems.But in TgAb-positive serum samples , the average recoveries in Roche and Beckman systems were 88.59% (35.85% -141.53%) and 95.77% (36.48% -131.78%) respectively, and 12.63%(24/190) and 13.68%(26/190) samples displayed a recovery less than 80%.The recovery rate of confirmed serum Tg showed a significantly negative correlation with elevated TgAb concentration , with r=-0.239 (P=0.001) in Roche and r=-0.251 (P<0.001) in Beckman.Conclusions TgAb-positive serum, especially with high concentration of TgAb , significantly interfered the measurement of Tg .Thus, serum TgAb should be determined together with serum Tg to explore whether there was an interference .To avoid misdiagnosis and inappropriate therapy , clinician should be informed once serum TgAb displayed positive.

Objective To construct and identify the recombinant vector pcDNA3. 1 (-) B/myc-BRMS 1 carrying breast-cancer metastasis suppressor 1 (BRMS 1) which can express in eukaryote cells and which will provide the basis for further researching the mechanisms of metastasis suppression and working on cancer metastasis gene ther-apy. Methods To isolate total RNA from MCF - 7 cells and design a pair of primers, and coding sequence of aRMS 1 cDNA were amplified from human breast cancer cells MCF -7 by reverse transcription-polymerase chain reaction (RT-PCR). Then the product was inserted to the PcDNA3. 1/myc-His (-) B plasmid. The recombined pcDNA3. 1 (-)B/myc-BRMS1 was identified by gene sequence analysis,then recombinants was transfected into HEK-293 cells and was identified by Western blot. Results The recombinant of pcDNA3.1 (-) B/myc-BRMS1 was structurally confirmed by analysis of sequencing. The inserted fragment in the vector was in the right direction and its sequence was structurally confirmed to be consistent with CDS sequence of human BRMSI cDNA that of the published data. GenBank, [AF159141]. The recombinants was transfected into HEK-293 cells ,then the cells expressed protein tagged c-myc identified by Western blot indicated it can express in eukaryote cells. Conclusion cDNA of human BRMS1 can be successfully cloned and inserted into Eukaryote-expression vector. The newly constructed vector may serve as the potential tool to conduct further comprehensive experiments in future on BRMS1 function and on gene therapy.

Objective To discuss the method of preventing injury of recurrent laryngeal nerve in thyroid surgery. Methods 362 cases of thyroid operation in our hospital were retrospectively studied, which consists of 188 cases with single lateral partial or total thyroidectomy, 127 with double lateral partial thyroidectomy,31 with single lateral total thyroidectomy plus isthmectomy, 16 with thyroid carcinoma radical correction. Results There was no injury,except for 3 cases damage to validate the recurrent laryngeal nerve in thyroid surgery,which were fully recov- ered after treatment. Conclusion To dissect recurrent laryngeal nerve routinely in inthyroid surgery can prevent and reduce the injury.