Genomic Models of Short-Term Exposure Accurately Predict Long-Term Chemical Carcinogenicity and Identify Putative Mechanisms of Action

Background

Despite an overall decrease in incidence of and mortality from cancer, about 40% of Americans will be diagnosed with the disease in their lifetime, and around 20% will die of it. Current approaches to test carcinogenic chemicals adopt the 2-year rodent bioassay, which is costly and time-consuming. As a result, fewer than 2% of the chemicals on the market have actually been tested. However, evidence accumulated to date suggests that gene expression profiles from model organisms exposed to chemical compounds reflect underlying mechanisms of action, and that these toxicogenomic models could be used in the prediction of chemical carcinogenicity.

Results

In this study, we used a rat-based microarray dataset from the NTP DrugMatrix Database to test the ability of toxicogenomics to model carcinogenicity. We analyzed 1,221 gene-expression profiles obtained from rats treated with 127 well-characterized compounds, including genotoxic and non-genotoxic carcinogens. We built a classifier that predicts a chemical''s carcinogenic potential with an AUC of 0.78, and validated it on an independent dataset from the Japanese Toxicogenomics Project consisting of 2,065 profiles from 72 compounds. Finally, we identified differentially expressed genes associated with chemical carcinogenesis, and developed novel data-driven approaches for the molecular characterization of the response to chemical stressors.

Conclusion

Here, we validate a toxicogenomic approach to predict carcinogenicity and provide strong evidence that, with a larger set of compounds, we should be able to improve the sensitivity and specificity of the predictions. We found that the prediction of carcinogenicity is tissue-dependent and that the results also confirm and expand upon previous studies implicating DNA damage, the peroxisome proliferator-activated receptor, the aryl hydrocarbon receptor, and regenerative pathology in the response to carcinogen exposure.     

How citizen science could improve species distribution models and their independent assessment

1. Species distribution models (SDM) have been increasingly developed in recent years, but their validity is questioned. Their assessment can be improved by the use of independent data, but this can be difficult to obtain and prohibitive to collect. Standardized data from citizen science may be used to establish external evaluation datasets and to improve SDM validation and applicability.
2. We used opportunistic presence‐only data along with presence–absence data from a standardized citizen science program to establish and assess habitat suitability maps for 9 species of amphibian in western France. We assessed Generalized Additive and Random Forest Models’ performance by (1) cross‐validation using 30% of the opportunistic dataset used to calibrate the model or (2) external validation using different independent datasets derived from citizen science monitoring. We tested the effects of applying different combinations of filters to the citizen data and of complementing it with additional standardized fieldwork.
3. Cross‐validation with an internal evaluation dataset resulted in higher AUC (Area Under the receiver operating Curve) than external evaluation causing overestimation of model accuracy and did not select the same models; models integrating sampling effort performed better with external validation. AUC, specificity, and sensitivity of models calculated with different filtered external datasets differed for some species. However, for most species, complementary fieldwork was not necessary to obtain coherent results, as long as the citizen science data were strongly filtered.
4. Since external validation methods using independent data are considered more robust, filtering data from citizen sciences may make a valuable contribution to the assessment of SDM. Limited complementary fieldwork with volunteer''s participation to complete ecological gradients may also possibly enhance citizen involvement and lead to better use of SDM in decision processes for nature conservation.

     

A Novel Geriatric Screening Tool in Older Patients with Cancer: The Korean Cancer Study Group Geriatric Score (KG)-7

Geriatric assessment (GA) is resource-consuming, necessitating screening tools to select appropriate patients who need full GA. The objective of this study is to design a novel geriatric screening tool with easy-to-answer questions and high performance objectively selected from a large dataset to represent each domain of GA. A development cohort was constructed from 1284 patients who received GA from May 2004 to April 2007. Items representing each domain of functional status, cognitive function, nutritional status, and psychological status in GA were selected according to sensitivity (SE) and specificity (SP). Of the selected items, the final questions were chosen by a panel of oncologists and geriatricians to encompass most domains evenly and also by feasibility and use with cancer patients. The selected screening questions were validated in a separate cohort of 98 cancer patients. The novel screening tool, the Korean Cancer Study Group Geriatric Score (KG)-7, consisted of 7 items representing each domain of GA. KG-7 had a maximal area under the curve (AUC) of 0.93 (95% confidence interval (CI) 0.92−0.95) in the prediction of abnormal GA, which was higher than that of G-8 (0.87, 95% CI 0.85–0.89) within the development cohort. The cut-off value was decided at ≤ 5 points, with a SE of 95.0%, SP of 59.2%, positive predictive value (PPV) of 85.3%, and negative predictive value (NPV) of 82.6%. In the validation cohort, the AUC was 0.82 (95% CI 0.73−0.90), and the SE, SP, PPV, and NPV were 89.5%, 48.6%, 77.3%, and 75.0%, respectively. Furthermore, patients with higher KG-7 scores showed significantly longer overall survival (OS) in the development and validation cohorts. In conclusions, the KG-7 showed high SE and NPV to predict abnormal GA. The KG-7 also predicted OS. Given the results of our studies, the KG-7 could be used effectively in countries with high patient burden and low resources to select patients in need of full GA and intervention.     

基于X线图像的膝关节周围原发性骨肿瘤辅助诊断的机器学习模型研究

摘要 目的：开发机器学习模型,并评估其在膝关节周围原发性骨肿瘤诊断方面的准确性。方法：本文将深度卷积神经网络（DCNN）这一深度学习方法应用于膝关节X线图像的影像组学分析,探讨其辅助诊断膝关节周围原发性骨肿瘤的临床价值。结果：该深度学习模型在区分正常与肿瘤影像方面展现出优异的诊断准确性,使用DCNN模型进行5轮测试的总体准确性为（99.8±0.4）%,而阳性预测值和阴性预测值分别为（100.0±0.0）%和（99.6±0.8）%,各个数据集的曲线下面积（AUC）分别为0.99、1.00、1.00、1.0和1.0,平均AUC为（0.998±0.004）;进一步使用DCNN模型进行了10轮测试显示其在区分良性与恶性骨肿瘤方面的总体准确性为（71.2±1.6）%,且达到了强阳性预测值（91.9±8.5）%,各个数据集的AUC分别为0.63、0.63、0.58、0.69、0.55、0.63、0.54、0.57、0.73、0.63,平均AUC为（0.62±0.06）。结论：本文是首个将人工智能技术应用于骨肿瘤诊断的X线图像影像组学分析方面的研究,人工智能影像组学模型能够帮助医生自动地快速筛查骨肿瘤,确定良性或恶性肿瘤时,阳性预测值较高。     

Evaluation of computational techniques for predicting non-synonymous single nucleotide variants pathogenicity

The human genetic diseases associated with many factors, one of these factors is the non-synonymous Single Nucleotide Variants (nsSNVs) cause single amino acid change with another resulting in protein function change leading to disease. Many computational techniques have been released to expect the impacts of amino acid alteration on protein function and classify mutations as pathogenic or neutral. Here in this article, we assessed the performance of eight techniques; FATHMM, SIFT, Provean, iFish, Mutation Assessor, PANTHER, SNAP2, and PON- P2 using a VaribenchSelectedPure dataset of 2144 pathogenic variants and 3777 neutral variants extracted from the free standard database “Varibench.” The first five techniques achieve (45.60–83.75) % specificity, (52.64–94.13) % sensitivity, (51.00–88.90) % AUC, and (49.76–88.24) % ACC on whole dataset, while all eight techniques achieve (36.54–77.88) % specificity, (50.00–75.00) % sensitivity, (51.00–76.40) % AUC, and (25.00–77.78) % ACC on random sample dataset. We also created a Meta classifier (CSTJ48) that combines FATHMM, iFish, and Mutation Assessor. It registers 96.33% specificity, 86.07% sensitivity, 91.20% AUC, and 91.89 ACC. By comparing the results, it's clear that FATHMM gives the highest performance over the seven individual techniques, where it achieves 83.75% and 77.88% specificity, 94.13%, and 75.00% sensitivity, 88.90% and 76.40% AUC, and 88.24% and 77.78% ACC on whole and random sample dataset, respectively. Also, the launched Meta classifier (CSTJ48) is outperforming over all the eight individual tools that compared here.     

Serum miR-1297: a promising diagnostic biomarker in esophageal squamous cell carcinoma

We aimed to value the diagnostic potential of serum miR-1297 in esophageal squamous cell cancer (ESCC). Its expression level was detected in 156 pairs of patients with ESCC and healthy volunteers using quantitative real-time polymerase chain reaction (qRT-PCR) method. It was statistically decreased in ESCC patients compared with healthy controls. AUC based on serum miR-1297 was 0.840?±?0.035 in discovery group and 0.837?±?0.034 in validation group. Further analysis on early-stage patients revealed that the AUC was 0.819?±?0.053 in discovery group and 0.814?±?0.044 in validation group. Its sensitivity and specificity were promising. In conclusion, serum miR-1297 can serve as an ideal indicator for the diagnosis of ESCC.     

An Improved Computational Prediction Model for Lysine Succinylation Sites Mapping on Homo sapiens by Fusing Three Sequence Encoding Schemes with the Random Forest Classifier

Background Lysine succinylation is one of the reversible protein post-translational modifications (PTMs), which regulate the structure and function of proteins. It plays a significant role in various cellular physiologies including some diseases of human as well as many other organisms. The accurate identification of succinylation site is essential to understand the various biological functions and drug development.Methods In this study, we developed an improved method to predict lysine succinylation sites mapping on Homo sapiens by the fusion of three encoding schemes such as binary, the composition of k-spaced amino acid pairs (CKSAAP) and amino acid composition (AAC) with the random forest (RF) classifier. The prediction performance of the proposed random forest (RF) based on the fusion model in a comparison of other candidates was investigated by using 20-fold cross-validation (CV) and two independent test datasets were collected from two different sources.Results The CV results showed that the proposed predictor achieves the highest scores of sensitivity (SN) as 0.800, specificity (SP) as 0.902, accuracy (ACC) as 0.919, Mathew correlation coefficient (MCC) as 0.766 and partial AUC (pAUC) as 0.163 at a false-positive rate (FPR) = 0.10 and area under the ROC curve (AUC) as 0.958. It achieved the highest performance scores of SN as 0.811, SP as 0.902, ACC as 0.891, MCC as 0.629 and pAUC as 0.139 and AUC as 0.921 for the independent test protein set-1 and SN as 0.772, SP as 0.901, ACC as 0.836, MCC as 0.677 and pAUC as 0.141 at FPR = 0.10 and AUC as 0.923 for the independent test protein set-2. It also outperformed all the other existing prediction models.Conclusion The prediction performances as discussed in this article recommend that the proposed method might be a useful and encouraging computational resource for lysine succinylation site prediction in the case of human population.     

Deep learning for predicting subtype classification and survival of lung adenocarcinoma on computed tomography

ObjectivesThe subtype classification of lung adenocarcinoma is important for treatment decision. This study aimed to investigate the deep learning and radiomics networks for predicting histologic subtype classification and survival of lung adenocarcinoma diagnosed through computed tomography (CT) images.MethodsA dataset of 1222 patients with lung adenocarcinoma were retrospectively enrolled from three medical institutions. The anonymised preoperative CT images and pathological labels of atypical adenomatous hyperplasia, adenocarcinoma in situ, minimally invasive adenocarcinoma, invasive adenocarcinoma (IAC) with five predominant components were obtained. These pathological labels were divided into 2-category classification (IAC; non-IAC), 3-category and 8-category. We modeled the classification task of histological subtypes based on modified ResNet-34 deep learning network, radiomics strategies and deep radiomics combined algorithm. Then we established the prognostic models in lung adenocarcinoma patients with survival outcomes. The accuracy (ACC), area under ROC curves (AUCs) and C-index were primarily performed to evaluate the algorithms.ResultsThis study included a training set (n = 802) and two validation cohorts (internal, n = 196; external, n = 224). The ACC of deep radiomics algorithm in internal validation achieved 0.8776, 0.8061 in the 2-category, 3-category classification, respectively. Even in 8 classifications, the AUC ranged from 0.739 to 0.940 in internal set. Further, we constructed a prognosis model that C-index was 0.892(95% CI: 0.846–0.937) in internal validation set.ConclusionsThe automated deep radiomics based triage system has achieved the great performance in the subtype classification and survival predictability in patients with CT-detected lung adenocarcinoma nodules, providing the clinical guide for treatment strategies.     

The diagnostic value of serum DSA-TRF in hepatocellular carcinoma

TRF is a glycoprotein mainly secreted by hepatocytes, The aim of this study was to explore the diagnostic value of aberrant glycosylated serum transferrin (TRF) especially containing multi-antennary glycans in hepatocellular carcinoma (HCC).A total of 581 subjects including HCC patients, liver cirrhosis (LC) patients, chronic hepatitis (CHB) patients and healthy controls (HC) were recruited. All the subjects were randomly assigned to training group (n?=?411) and validation group (n?=?170). We firstly analyzed the serum protein N-glycome profiling of HCC, LC, and HC by DNA sequencer–assisted fluorophore-assisted carbohydrate electrophoresis (DSA-FACE) technology. We established a lectin-antibody sandwich ELISA (Lectin-ELISA) method to detect multi-antennary glycans-contained TRF (DSA-TRF) in serum, in which Datura stramonium Agglutinin (DSA) was used for specific recognition. Levels of serum DSA-TRF and TRF were analyzed respectively. The diagnostic efficacies of DSA-TRF and TRF of differentiating HCC patients from CHB, LC patients and HC within the training group were evaluated using receiver operating characteristic (ROC) curve and tested in the validation group.The result found that in training group, serum TRF and DSA-TRF levels differed significantly between HCC (1.86?±?0.50, g/L, 0.285?±?0.06), CHB?+?LC (2.39?±?0.74, g/L, 0.189?±?0.07) and HC (1.92?±?0.69, g/L, 0.249?±?0.09) (HCC vs. CHB?+?LC, P?<?0.001; HCC vs. HC, P?<?0.001; CHB?+?LC vs. HC, P?<?0.001). The area under the ROC curve (AUC) of DSA-TRF was significantly superior to AFP (0.880, 95%CI: 0.834–0.925 vs. 0.776, 95%CI: 0.725–0.827, P?=?0.003) in differentiating HCC from CHB?+?LC. The AUC of diagnostic model GlycoTRF1 (0.981, 95%CI: 0.969–0.993) was higher than DSA-TRF and AFP alone (P<0.001) in differentiating HCC from CHB?+?LC, which was verified in validation group.The results indicated that the serum DSA-TRF might serve as a potential glycan biomarker for distinguishing HCC from CHB and LC.

     

Testing the generalizability of cfDNA fragmentomic features across different studies for cancer early detection

cfDNA fragmentomic features have been used in cancer detection models; however, the generalizability of the models needs to be tested. We proposed a type of cfDNA fragmentomic feature named chromosomal arm-level fragment size distribution (ARM-FSD), evaluated and compared its performance and generalizability for lung cancer and pan-cancer detection with existing cfDNA fragmentomic features (as reference) by using cohorts from different institutions. The ARM-FSD lung cancer model outperformed the reference model by ∼10% when being tested by two external cohorts (AUC: 0.97 vs. 0.86; 0.87 vs. 0.76). For pan-cancer detection, the performance of the ARM-FSD based model is consistently higher than the reference (AUC: 0.88 vs. 0.75, 0.98 vs. 0.63) in a pan-cancer and a lung cancer external validation cohort, indicating that ARM-FSD model produces stable performance across multiple cohorts. Our study reveals ARM-FSD based models have a higher generalizability, and highlights the necessity of cross-study validation for predictive model development.     

A deep learning classifier for digital breast tomosynthesis

PurposeTo develop a computerized detection system for the automatic classification of the presence/absence of mass lesions in digital breast tomosynthesis (DBT) annotated exams, based on a deep convolutional neural network (DCNN).Materials and MethodsThree DCNN architectures working at image-level (DBT slice) were compared: two state-of-the-art pre-trained DCNN architectures (AlexNet and VGG19) customized through transfer learning, and one developed from scratch (DBT-DCNN). To evaluate these DCNN-based architectures we analysed their classification performance on two different datasets provided by two hospital radiology departments. DBT slice images were processed following normalization, background correction and data augmentation procedures. The accuracy, sensitivity, and area-under-the-curve (AUC) values were evaluated on both datasets, using receiver operating characteristic curves. A Grad-CAM technique was also implemented providing an indication of the lesion position in the DBT slice.Results Accuracy, sensitivity and AUC for the investigated DCNN are in-line with the best performance reported in the field. The DBT-DCNN network developed in this work showed an accuracy and a sensitivity of (90% ± 4%) and (96% ± 3%), respectively, with an AUC as good as 0.89 ± 0.04. A k-fold cross validation test (with k = 4) showed an accuracy of 94.0% ± 0.2%, and a F1-score test provided a value as good as 0.93 ± 0.03. Grad-CAM maps show high activation in correspondence of pixels within the tumour regions.Conclusions We developed a deep learning-based framework (DBT-DCNN) to classify DBT images from clinical exams. We investigated also a possible application of the Grad-CAM technique to identify the lesion position.     

Computed tomography-based radiomic model at node level for the prediction of normal-sized lymph node metastasis in cervical cancer

PurposeRadiomic models have been demonstrated to have acceptable discrimination capability for detecting lymph node metastasis (LNM). We aimed to develop a computed tomography–based radiomic model and validate its usefulness in the prediction of normal-sized LNM at node level in cervical cancer.MethodsA total of 273 LNs of 219 patients from 10 centers were evaluated in this study. We randomly divided the LNs from the 2 centers with the largest number of LNs into the training and internal validation cohorts, and the rest as the external validation cohort. Radiomic features were extracted from the arterial and venous phase images. We trained an artificial neural network (ANN) to develop two single-phase models. A radiomic model reflecting the features of two-phase images was also built for directly predicting LNM in cervical cancer. Moreover, four state-of-the-art methods were used for comparison. The performance of all models was assessed using the area under the receiver operating characteristic curve (AUC).ResultsAmong the models we built, the models combining the features of two phases surpassed the single-phase models, and the models generated by ANN had better performance than the others. We found that the radiomic model achieved the highest AUCs of 0.912 and 0.859 in the training and internal validation cohorts, respectively. In the external validation cohort, the AUC of the radiomic model was 0.800.ConclusionWe constructed a radiomic model that exhibited great ability in the prediction of LNM. The application of the model could optimize clinical staging and decision-making.     

Structure based design of heat shock protein 90 inhibitors acting as anticancer agents

Structure based drug design (SBDD) was used to discover heat shock protein 90 (HSP90) inhibitors useful in the treatment of cancer. By using the crystal structure of HSP90-ligand complex (1uyi), a docking model was prepared and was validated by external dataset containing known HSP90 inhibitors. This validated model was then used to virtually screen commercial databases, selected hits of which were bought and sent for real biological evaluation. Further as an alternative method, pharmacophores were generated using crystal structure conformations of ligands in HSP90 complexes (1uyi and 2bz5) and where used for virtual screening. Both cases yielded several hits containing novel scaffolds, particularly compound KHSP8 showed an IC(50) value of 0.902 μM in case of colon cancer (HT29), which is comparable to doxorubicin (0.828 μM). These compounds were being now used as leads for constructing small molecular libraries to get compounds with favourable pharmacokinetics and drug like properties.     

Reactivity of pyruvic acid and its derivatives towards reactive oxygen species

Pyruvic acid and its derivatives occurring in most biological systems are known to exhibit several pharmacological properties, such as anti‐inflammatory, neuroprotective or anticancer, many of which are suggested to originate from their antioxidant and free radical scavenger activity. The therapeutic potential of these compounds is a matter of particular interest, due to their mechanisms of action, particularly their possible antioxidant behaviour. Here, we report the results of a study of the effect of pyruvic acid (PA), ethyl pyruvate (EP) and sodium pyruvate (SP) on reactions generating reactive oxygen species (ROS), such as superoxide anion radicals, hydroxyl radicals and singlet oxygen, and their total antioxidant capacity. Chemiluminescence (CL) and spectrophotometry techniques were employed. The pyruvate analogues studied were found to inhibit the CL signal arising from superoxide anion radicals in a dose‐dependent manner with IC₅₀ = 0.0197 ± 0.002 mM for EP and IC₅₀ = 69.2 ± 5.2 mM for PA. These compounds exhibited a dose‐dependent decrease in the CL signal of the luminol + H₂O₂ system over the range 0.5–10 mM with IC₅₀ values of 1.71 ± 0.12 mM for PA, 3.85 ± 0.21 mM for EP and 22.91 ± 1.21 mM for SP. Furthermore, these compounds also inhibited hydroxyl radical‐dependent deoxyribose degradation in a dose‐dependent manner over the range 0.5–200 mM, with IC₅₀ values of 33.2 ± 0.3 mM for SP, 116.1 ± 6.2 mM for EP and 168.2 ± 6.2 mM for PA. All the examined compounds also showed antioxidant capacity when estimated using the ferric–ferrozine assay. The results suggest that the antioxidant activities of pyruvate derivatives may reflect a direct effect on scavenging ROS and, in part, be responsible for their pharmacological actions. Copyright © 2015 John Wiley & Sons, Ltd.     

南海陆坡沉积物细菌丰度预测

分别利用参数模型和无参数估计法预测南海陆坡沉积物柱MD05-2896中的细菌丰度.基于非培养的PCR-RFLP的16SrRNA基因分子技术,扩增了沉积物柱中的细菌16S rRNA基因序列,并构建16S rRNA基因文库.系统发育分析表明16S rRNA基因文库中,大多数序列属于17个已知的“门”.分别以99％、97％、90％和80％序列一致性作为分类单元分界点,将16SrRNA基因序列组群为分类单元.使用逆高斯分布模型、对数正态分布模型、负二项式分布模型、帕雷托分布模型、双指数分布模型以及ACE、ACE-1等估计方法预测不同分类单元分类水平下的细菌丰度.结果表明在“种”级分类水平上,负二项式分布为最优估计模型,估计细菌丰度为244±10(SE).不过,受实验条件的限制,该估计值可能偏低.     

Age‐ and temperature‐dependent oviposition model of Neoseiulus californicus (McGregor) (Acari: Phytoseiidae) with Tetranychus urticae as prey

In this study, we developed an oviposition model of Neoseiulus californicus (McGregor) with Tetranychus urticae Koch as prey. To obtain data for the model, we investigated the longevity, fecundity and survivorship of adult female N. californicus at six constant temperatures (16, 20, 24, 28, 32 and 36°C), 60–70% RH and a photoperiod of 16 : 8 (L : D) h. Longevity (average ± SE) decreased as temperature increased and was longest at 16°C (46.7 ± 5.25 days) and shortest at 36°C (12.8 ± 0.75 days). Adult developmental rate (1/average longevity) was described by the Lactin 1 model (r² = 0.95). The oviposition period (average±SE) was also longest at 16°C (29.8 ± 2.93 days) and shortest at 36°C (6.7 ± 0.54 days). Fecundity (average±SE) was greatest at 24°C (43.8 ± 3.23 eggs) and lowest at 36°C (15.9 ± 1.50 eggs). The oviposition model comprised temperature‐dependent fecundity, age‐specific cumulative oviposition rate and age‐specific survival rate functions. The temperature‐dependent fecundity was best described by an exponential equation (r² = 0.81). The age‐specific cumulative oviposition rate was best described by the three‐parameter Weibull function (r² = 0.96). The age‐specific survival rate was best described by a reverse sigmoid function (r² = 0.85).     

Seasonal variability in survivorship of a cooperatively breeding tropical passerine

Survival of tropical passerines is thought to be higher than those in northern temperate regions, but relatively few tropical studies have addressed this issue, particularly in tropical Asia. We examined factors that may have influenced the survival rate of a cooperatively breeding bird, the puff-throated bulbul (Alophoixus pallidus), in an evergreen forest in northeastern Thailand. These factors included year, season (breeding and non-breeding), sex, and presence of helper(s) in a family group. We present evidence of breeding season-dependent survival in a tropical passerine using an information theoretic approach based on both mark-recapture and resighting data collected during 6 years of study. Based on colour-banded adults the annual survival rate did not vary significantly among years (average = 0.85 ± 0.02 SE). The mean lifespan (MLS) for the population was 6.22 ± 4.38 SE years. Survivorship was lower during the breeding season (0.89 ± 0.02 SE) than during the non-breeding season (0.96 ± 0.02 SE). The MLS of males and females was 6.70 ± 7.73 SE and 5.87 ± 4.88 SE years, respectively. The annual survival rate we observed was high compared to the estimates of other tropical and temperate passerines, possibly due to the relatively stable climatic conditions in tropical latitudes and puff-throated bulbuls being generalists that exploit a wide range of food resources both in space and time.