Debris flow volume prediction model based on back propagation neural network optimized by improved whale optimization algorithm.

Debris flow is a sudden natural disaster in mountainous areas, which seriously threatens the lives and property of nearby residents. Therefore, it is necessary to predict the volume of debris flow accurately and reliably. However, the predictions of back propagation neural networks are unstable and inaccurate due to the limited dataset. In this study, the Cubic map optimizes the initial population position of the whale optimization algorithm. Meanwhile, the adaptive weight adjustment strategy optimizes the weight value in the shrink-wrapping mechanism of the whale optimization algorithm. Then, the improved whale optimization algorithm optimizes the final weights and thresholds in the back propagation neural network. Finally, to verify the performance of the final model, sixty debris flow gullies caused by earthquakes in Longmenshan area are selected as the research objects. Through correlation analysis, 4 main factors affecting the volume of debris flow are determined and inputted into the model for training and prediction. Four methods (support vector machine regression, XGBoost, back propagation neural network optimized by artificial bee colony algorithm, back propagation neural network optimized by grey wolf optimization algorithm) are used to compare the prediction performance and reliability. The results indicate that loose sediments from co-seismic landslides are the most important factor influencing the flow of debris flows in the earthquake area. The mean absolute percentage error, mean absolute error and R2 of the final model are 0.193, 29.197 × 104 m3 and 0.912, respectively. The final model is more accurate and stable when the dataset is insufficient and under complexity. This is attributed to the optimization of WOA by Cubic map and adaptive weight adjustment. In general, the model of this paper can provide reference for debris flow prevention and machine learning algorithms.

Circulating lipids, lipid-lowering drug targets, and breast cancer risk: Comprehensive evidence from Mendelian randomization and summary data-based Mendelian randomization.

BACKGROUND: Breast cancer (BC) is the most common and fatal cancer among women, yet the causal relationship between circulating lipids, lipid-lowering drugs, and BC remains unclear. METHODS: Mendelian randomization (MR) and summary data-based MR (SMR) analysis are used to explore the causal relationship between plasma lipids, lipid-lowering drug targets, and BC. RESULTS: The result of MR suggested that per mg/dL higher levels of LDL-C (OR = 1.045, FDR = 0.023), HDL-C (OR = 1.079, FDR = 0.003), TC (OR = 1.043, FDR = 0.026), and APOA-I (OR = 1.085, FDR = 2.64E-04) were associated with increased BC risk, while TG was associated with reduced BC risk (OR = 0.926, FDR = 0.003). Per mg/dL higher levels of HDL-C (OR = 1.080, FDR = 0.011) and APOA-I (OR = 1.083, FDR = 0.002) were associated with increased ER+BC risk, while TG was associated with reduced ER+BC risk (OR = 0.909, FDR = 0.002). For every per 1 mg/dL decrease in LDL, HMGCR (OR: 0.839; FDR = 0.016), NPC1L1 (OR: 0.702; FDR = 0.004), and PCSK9 (OR: 0.916; FDR = 0.026) inhibition were associated with reduced BC risk, whereas CETP inhibition (OR: 1.194; FDR = 0.026) was associated with increased BC risk. For every per 1 mg/dL decrease in LDL, HMGCR (OR: 0.822; FDR = 0.023), NPC1L1 (OR: 0.633; FDR = 2.37E-03), and APOB inhibition (OR: 0.816; FDR = 1.98E-03) were associated with decreased ER-BC risk, while CETP inhibition (OR: 1.465; FDR = 0.011) was associated with increased ER-BC risk. SMR analysis indicated that HMGCR was associated with increased BC risk (OR: 1.112; p = 0.044). CONCLUSION: Lipids are associated with the BC risk, and lipid-lowering drugs targets HMGCR, NPC1L1, PCSK9, and APOB may be effective strategies for preventing BC. However, lipid-lowering drugs target CETP may potentially increase BC risk.

Tunnel collapse risk assessment based on improved quantitative theory III and EW-AHP coupling weight.

It is a multi-criteria decision issue to conduct a risk assessment of the tunnel. In this paper, a tunnel collapse risk assessment model based on the improved theory of quantification III and the fuzzy comprehensive evaluation method is proposed. According to the geological conditions and the construction disturbance classification method, the evaluation factors are selected, and the tunnel collapse risk level is divided into 5 levels according to the principle of maximum membership degree. The three groups of scores with the largest correlation ratio are calculated by the theory of quantification III to form the X, Y, and Z axes of the spatial coordinate system, The spatial distance of each evaluation factor is optimized by the Kendall correlation coefficient combined with the empirical formula, so that it can be used to judge the probability of the occurrence of the evaluation factor; taking the coupling of the objective entropy weight method (EW) and the subjective analytic hierarchy process (AHP) as the weight. Finally, the fuzzy comprehensive evaluation method is used to determine the possibility classification of tunnel collapse. Taking the Ka-Shuang water diversion tunnel as a case study, the comparison between the evaluation results of 10 tunnel samples and the status quo of the actual engineering area verifies the reliability of the method.

[Allometry of scales in Chinese pangolins (Manis pentadactyla) and Malayan pangolins (Manis javanica) and application in judicial expertise].

Pangolins are unique mammals in that they possess scales that serve a protective biological function. As an important raw material of traditional medicine, illegal trades of these scales are frequent and difficult to investigate or prosecute. We used allometric models of dry weight of scales to compare 35 Chinese pangolins (Manis pentadactyla) and 119 Malayan pangolins (Manis javanica). Our results showed that the dry weight of scales increases significantly faster with the length of head and body in Malayan pangolins (P=0.005), while dry weight of scales is positive (slope=3.725) in Malayan pangolins but isometric (slope=3.105) in Chinese pangolins. The differences in morphology between these species may reflect an evolutionary adaptation to different environments; Malayan pangolins in tropical regions appear to suffer from greater predation pressure than Chinese pangolins in temperate regions. We advise the conversion standards between dry weight of scales and number of individuals as 573.47 g in Chinese pangolins and 360.51 g in Malayan pangolins respectively, and when two are mixed together, average above two parameters of the median at 466.99 g. We propose these measurements be used as judicial evidences in forensic identification of related cases.