Single nueleotide polymorphisms (SNPs) are the most common genetic variants in human genome.Candidate gene,genome-wide association studies (GWASs) and exome sequencing which base on SNPs have made a great progress in identifying cancer susceptibility.The development and application of high resolutions in SNPs has played an important role in clarifying the mechanism,prevention,diagnosis and targeted therapy in cancers.

With the emergence of high-throughput technology and massive toxicology data,toxicology research has entered the era of big data.How to efficiently integrate existingtoxicological data,clarify the toxic effects of chemicals,and use these patterns to providenew information,in order to achieve effi-cient prediction of the toxicity of new chemicalsubstances,is one of the cutting-edge issues in toxicology.In view of the high cost,low throughput and difficulty in revealing the mechanism information of tradi-tional chemical toxicity testing methods,high throughput prediction models are urgently needed.Machine learning methods have been applied to toxicity testing,such as supervised learning models,unsupervised learning models,deep learning models,reinforcement learning models,and transfer learning models.Chemical characteristic data commonly used in machine learning models include chemical structure data,text data,toxicological genome data and image data.There is huge potential for applying machine learning to toxicity testing and machine learning methods have made some prog-ress.However,current research focuses on the processing of data and development of models,which has failed to produce a widely used and accepted method.In addition,the prediction accuracy of machine learning models is not only dependent on algorithms,but also affected by data quality,and the mutual promotion and development of algorithms and data quality remains a big challenge.In short,data processing and model construction in the field of toxicology require interdisciplinary cooperation and technological innovation.With the increasing perfection of toxicology databases and the continuous optimization of various model algorithms,the toxicity prediction of new chemicals based on machine learning models will become increasingly efficient and accurate,playing an important role in ensuring human health and environmental safety.

Objective To detect the association between rs4646999 polymorphisms in the promoter region of the c-Jun and the prognosis of sporadic colorectal cancer. Methods rs4646999-673C>T genetypes were deter-mined by Taqman-MGB probes in 436 colorectal cancer cases. The survival curve was analyzed by Kaplan-Meier analysis and Cox regression.Western blot was used to analyze the expression levels of c-Jun protein in different gen-otypes. Results Univariate analysis showed that the cumulative survival rate of patients with rs4646999TT geno-type was significantly higher than that of patients with CT and CC genotype. Multivariate Cox regression analysis showed that the differentiation,lymph node metastasis,distant metastasis,TNM stage and rs4646999 genetypes were prognostic factors.Compared with the carriers of TT genotype,CT/CC complex genotypes were associated with poor prognosis of colorectal cancer(P<0.05).Protein expression analysis showed that the expression of c-Jun pro-tein in CC genotype was increased.In contrast,the TT genotype was decreased.Conclusions This study provided the evidence that rs4646999-673C>T genetic variant in c-Jun promoter regions is associated with the poor survival prognosis of colorectal cancer,possibly by elevating the protein expression levels that appeared to up-regulate activ-ity of c-Jun thus tumorigenesis.

OBJECTIVETo explore the associations between the genetic variations in the SDC2 gene and overall survival and risk of radiation esophagitis in patients with esophageal squamous cell carcinoma (ESCC).

METHODSEleven functional haplotype-tagging single nucleotide polymorphisms (htSNPs) of SDC2 were genotyped in 296 ESCC patients who received radiotherapy alone, and had different response and esophagitis. The associations between genotypes and risk of esophagitis were measured by odds ratios (ORs) and 95% confidence intervals (CIs), adjusted for sex, age, tumor location, staging, radiotherapy mode and total radiation dose. The hazard ratios (HRs) were estimated using Cox proportional hazards regression model.

RESULTSThe median survival time (MST) of these patients was 14 months. Of them, 260 (87.8%) had died until the last date of follow-up of 30 June, 2014. Clinical stage (stage IV vs. stage II) and total radiation dose (≥ 60 Gy vs. < 60 Gy) influence the overall survival time of the patient significantly. Cox proportional hazards regression model analysis showed that the subjects with rs61599409 T allele had an decreased hazard ratio as compared with those with C allele (adjusted HR = 0.82, 95% CI, 0.66-1.02), but the difference was not statistically significant (P = 0.071). The rest 10 htSNPs were not associated with the overall survival of ESCC patients treated with radiotherapy. Among this set of patients, 160 (54.1%) suffered from radiation esophagitis. We found that rs17788084 A > T SNP in the 3'-untranslational region of SDC2 was associated with esophagitis risk, with the OR being 0.48 (95% CI = 0.28-0.85, P = 0.011) for the TA or TT genotype compared with the AA genotype.

CONCLUSIONSThese results suggest that rs17788084 genetic variation in SDC2 is associated with risk of radiation esophagitis and might serve as a potential biomarker for personalized radiotherapy of ESCC.

Alleles ; Carcinoma, Squamous Cell ; mortality ; pathology ; radiotherapy ; Esophageal Neoplasms ; mortality ; pathology ; radiotherapy ; Esophagitis ; genetics ; Genetic Variation ; Genotype ; Haplotypes ; Humans ; Odds Ratio ; Polymorphism, Single Nucleotide ; Proportional Hazards Models ; Radiation Injuries ; genetics ; Radiotherapy Dosage ; Risk ; Survival Analysis ; Syndecan-2 ; genetics ; Time Factors

model. Results The median survival time (MST) of these patients was 14 months. Of them, 260 (87.8%) had died until the last date of follow?up of 30 June, 2014. Clinical stage ( stage Ⅳ vs. stage Ⅱ) and total radiation dose (≥ 60 Gy vs. <60 Gy) influence the overall survival time of the patient significantly. Cox proportional hazards regression model analysis showed that the subjects with rs61599409 T allele had an decreased hazard ratio as compared with those with C allele ( adjusted HR=0.82, 95% CI, 0.66?1.02) , but the difference was not statistically significant ( P=0.071) . The rest 10 htSNPs were not associated with the overall survival of ESCC patients treated with radiotherapy. Among this set of patients, 160 ( 54. 1%) suffered from radiation esophagitis. We found that rs17788084 A > T SNP in the 3′?untranslational region of SDC2 was associated with esophagitis risk, with the OR being 0.48 (95% CI=0.28?0.85, P=0.011) for the TA or TT genotype compared with the AA genotype. Conclusions These results suggest that rs17788084 genetic variation in SDC2 is associated with risk of radiation esophagitis and might serve as a potential biomarker for personalized radiotherapy of ESCC.

Objective To explore the associations between genetic variations of DNA repair gene RAD52 and response to platinum?based chemotherapy of small cell lung cancer ( SCLC) , and to analyze the influencing factors on survival. Methods Nine haplotype?tagging single nucleotide polymorphisms ( htSNPs) of RAD52 were genotyped by Sequenom Mass ARRAY technology in 939 SCLC patients who received platinum?based chemotherapy, and had different response and survival time. The associations between genotypes and platinum?based chemotherapy response were analyzed by odds ratios ( ORs) and 95%confidence intervals ( CIs) , adjusted for sex, age, smoking, KPS, staging and chemotherapy regimens, by unconditional logistic regression model. The relative ratios ( RRs ) were estimated using Cox proportional hazards regression model. Results Among the 939 cases, 483 ( 51. 4%) cases received cis?platinum and etoposide treatment while others treated with carboplatin and etoposide. Six hundred and eighty two patients were chemotherapy responders in the study with a response rate of 72.6%. Patients were followed up to get their survival information. The median survival time ( MST) of these patients was 25 months. We found that rs10774474 SNP which located in the 5′?flanking region of RAD52 was significantly associated with chemotherapy response. Compared with the TT genotype, patients with TA and AA genotype had a worse chemotherapy response and increased risk of no?response ( P=0. 004 ) . Correlation analysis showed that patients with KPS >80 had a better chemotherapy response than those with KPS≤80 (P=0.001). The patients with extensive?stage had a worse chemotherapy response than those with limited?stage (P<0.001). Cox proportional hazards regression model analysis showed that nine htSNPs of RAD52 were not associated with the overall survival ( OS) of SCLC patients who received platinum?based chemotherapy. Age≤56, KPS>80, limited?stage, chemotherapy response and radiation therapy can remarkably prolong OS ( all P<0.05) . Conclusions These results suggest that RAD52 genetic polymorphism rs10774474 plays an important role in the response to platinum?based chemotherapy, and may be a potential genetic biomarker for SCLC personalized treatment.

model. Results The median survival time (MST) of these patients was 14 months. Of them, 260 (87.8%) had died until the last date of follow?up of 30 June, 2014. Clinical stage ( stage Ⅳ vs. stage Ⅱ) and total radiation dose (≥ 60 Gy vs. <60 Gy) influence the overall survival time of the patient significantly. Cox proportional hazards regression model analysis showed that the subjects with rs61599409 T allele had an decreased hazard ratio as compared with those with C allele ( adjusted HR=0.82, 95% CI, 0.66?1.02) , but the difference was not statistically significant ( P=0.071) . The rest 10 htSNPs were not associated with the overall survival of ESCC patients treated with radiotherapy. Among this set of patients, 160 ( 54. 1%) suffered from radiation esophagitis. We found that rs17788084 A > T SNP in the 3′?untranslational region of SDC2 was associated with esophagitis risk, with the OR being 0.48 (95% CI=0.28?0.85, P=0.011) for the TA or TT genotype compared with the AA genotype. Conclusions These results suggest that rs17788084 genetic variation in SDC2 is associated with risk of radiation esophagitis and might serve as a potential biomarker for personalized radiotherapy of ESCC.

Objective To explore the associations between genetic variations of DNA repair gene RAD52 and response to platinum?based chemotherapy of small cell lung cancer ( SCLC) , and to analyze the influencing factors on survival. Methods Nine haplotype?tagging single nucleotide polymorphisms ( htSNPs) of RAD52 were genotyped by Sequenom Mass ARRAY technology in 939 SCLC patients who received platinum?based chemotherapy, and had different response and survival time. The associations between genotypes and platinum?based chemotherapy response were analyzed by odds ratios ( ORs) and 95%confidence intervals ( CIs) , adjusted for sex, age, smoking, KPS, staging and chemotherapy regimens, by unconditional logistic regression model. The relative ratios ( RRs ) were estimated using Cox proportional hazards regression model. Results Among the 939 cases, 483 ( 51. 4%) cases received cis?platinum and etoposide treatment while others treated with carboplatin and etoposide. Six hundred and eighty two patients were chemotherapy responders in the study with a response rate of 72.6%. Patients were followed up to get their survival information. The median survival time ( MST) of these patients was 25 months. We found that rs10774474 SNP which located in the 5′?flanking region of RAD52 was significantly associated with chemotherapy response. Compared with the TT genotype, patients with TA and AA genotype had a worse chemotherapy response and increased risk of no?response ( P=0. 004 ) . Correlation analysis showed that patients with KPS >80 had a better chemotherapy response than those with KPS≤80 (P=0.001). The patients with extensive?stage had a worse chemotherapy response than those with limited?stage (P<0.001). Cox proportional hazards regression model analysis showed that nine htSNPs of RAD52 were not associated with the overall survival ( OS) of SCLC patients who received platinum?based chemotherapy. Age≤56, KPS>80, limited?stage, chemotherapy response and radiation therapy can remarkably prolong OS ( all P<0.05) . Conclusions These results suggest that RAD52 genetic polymorphism rs10774474 plays an important role in the response to platinum?based chemotherapy, and may be a potential genetic biomarker for SCLC personalized treatment.

Background Oral exposure to hexavalent chromium [Cr(VI)] can lead to gastrointestinal tumorigenesis in mice, and the mechanism is not yet clear. To predict health risk due to chemical exposure, data mining and computational toxicology analysis has become an important tool in toxicology research, which can help to elucidate mode of action (MOA) and identify key toxicity pathways. Objective This study aims to identify and evaluate key events in the MOA of oral Cr(VI) exposure. Methods Gene sets established from Comparative Toxicogenomics Database (CTD) and Gene Expression Omnibus (GEO) respectively were imported into Ingenuity® Pathway Analysis (IPA) software for pathway enrichment analysis and biological function analysis to identify potential key toxicity pathways of target organs/tissues toxicity of oral exposure to Cr(Ⅵ). Next, the weight of evidence (WOE) of the identified key toxicity pathways in the MOA of oral exposure to Cr(VI) was evaluated based on the modified Bradford Hill principle. Results A total of 54 pieces of literature related to oral Cr(VI) exposure were screened in CTD, among which 18 and 9 were related to liver and intestine with 125 and 272 corresponding genes, respectively. The pathway enrichment and biological function analysis results showed that liver and intestinal perturbation pathways were mainly related to cell stress and injury, cell cycle regulation, and apoptosis, indicating that Nrf2 pathway and AHR pathway might be the key toxicity pathways involved in the cytotoxic-mediated MOA. Meanwhile, the dose (≥170 mg·L⁻¹ sodium dichromate) and the time point (90 d) of the activation of Nrf2 pathway was similar to the emergence of crypt cell proliferation. It was proposed that Nrf2 pathway activation might be a key event for cytotoxic-mediated MOA of small intestinal tumors. The WOE results showed moderate validity of evidence in this hypothesis, with high validity of evidence for biological plausibility and dose-response manner. Conclusion Nrf2 pathway activation might be the key event in the cytotoxic-mediated MOA of small intestinal tumors induced by oral exposure to Cr(VI) via initiating or maintaining crypt cell proliferation.

Objective: To explore the associations between genetic variations of glutathione synthetase gene (GSS) and response to platinum-based chemotherapy of small cell lung cancer(SCLC), and to analyze the influencing factors on survival. Methods: Four haplotype-tagging single nucleotide polymorphisms (htSNPs) of GSS were genotyped by Sequenom MassARRAY methods in 903 SCLC patients who received platinum-based chemotherapy, and had different response and survival time. The associations between genotypes and platinum-based chemotherapy response were measured by odds ratios (OR) and 95% confidence intervals (CI), adjusted for sex, age, smoking, KPS, staging and chemotherapy regiments, by unconditional logistic regression model. The hazard ratios (HR) were estimated by Cox proportional hazards regression model. Results: Among the 903 patients, 462(51.2%) cases received cis-platinum and etoposide treatment while others were treated with carboplatin and etoposide. 656 patients were chemotherapy responders in the study with a response rate of 72.6%. Patients were followed up to get their survival information. The median survival time (MST) of these patients was 25.0 months.We found that rs725521 located in the 3′ near gene region of GSS was significantly associated with chemotherapy response. Compared with the T allele, patients with C allele had a worse chemotherapy response and an increased risk of no-responders (P=0.027). Rs7265992 and rs725521 of GSS were associated with the overall survival (OS) of SCLC patients who received platinum-based chemotherapy (HR=1.16, 95% CI=1.02-1.33, P=0.027; HR=1.17, 95% CI=1.05-1.31, P=0.006, respectively). The patients carrying 1 or 2 risk alleles and the patients carrying 3 or 4 risk alleles had worse MST than the patients without the rs7265992A and rs725521C risk alleles (24.0 and 22.0 versus 30.0 months), with the HR for death being 1.26 (95% CI=1.04-1.54) and with the HR of 1.52 (95%CI=1.18-1.97, P=0.001). Rs2025096 and rs2273684 were not associated with the OS of SCLC patients who received platinum-based chemotherapy. Age ≤ 56, KPS> 80, limited-stage, chemotherapy response and radiation therapy had a remarkably prolonged OS (all P<0.05). Conclusions These results suggest that GSS genetic polymorphism rs725521 plays an important role in the response to platinum-based chemotherapy, while rs7265992 and rs725521 have important effect on the prognosis of SCLC patients, which may be potential genetic biomarkers for personalized treatment of SCLC.