Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “logistic regression”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 199 records · Page 11

Multivariate Normal Tissue Complication Probability Modeling of Heart Valve Dysfunction in Hodgkin Lymphoma Survivors

Purpose: To establish a multivariate normal tissue complication probability (NTCP) model for radiation-induced asymptomatic heart valvular defects (RVD). Methods and Materials: Fifty-six patients treated with sequential chemoradiation therapy for Hodgkin lymphoma (HL) were retrospectively reviewed for RVD events. Clinical information along with whole heart, cardiac chambers, and lung dose distribution parameters was collected, and the correlations to RVD were analyzed by means of Spearman's rank correlation coefficient (Rs). For the selection of the model order and parameters for NTCP modeling, a multivariate logistic regression method using resampling techniques (bootstrapping) was applied. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC). Results: When we analyzed the whole heart, a 3-variable NTCP model including the maximum dose, whole heart volume, and lung volume was shown to be the optimal predictive model for RVD (Rs = 0.573, P<.001, AUC = 0.83). When we analyzed the cardiac chambers individually, for the left atrium and for the left ventricle, an NTCP model based on 3 variables including the percentage volume exceeding 30 Gy (V30), cardiac chamber volume, and lung volume was selected as the most predictive model (Rs = 0.539, P<.001, AUC = 0.83; and Rs = 0.557, P<.001, AUC = 0.82, respectively). The NTCP values increase as heart maximum dose or cardiac chambers V30 increase. They also increase with larger volumes of the heart or cardiac chambers and decrease when lung volume is larger. Conclusions: We propose logistic NTCP models for RVD considering not only heart irradiation dose but also the combined effects of lung and heart volumes. Our study establishes the statistical evidence of the indirect effect of lung size on radio-induced heart toxicity.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Learning epistatic polygenic phenotypes with Boolean interactions

Detecting epistatic drivers of human phenotypes is a considerable challenge. Traditional approaches use regression to sequentially test multiplicative interaction terms involving pairs of genetic variants. For higher-order interactions and genome-wide large-scale data, this strategy is computationally intractable. Moreover, multiplicative terms used in regression modeling may not capture the form of biological interactions. Building on the Predictability, Computability, Stability (PCS) framework, we introduce the epiTree pipeline to extract higher-order interactions from genomic data using tree-based models. The epiTree pipeline first selects a set of variants derived from tissue-specific estimates of gene expression. Next, it uses iterative random forests (iRF) to search training data for candidate Boolean interactions (pairwise and higher-order). We derive significance tests for interactions, based on a stabilized likelihood ratio test, by simulating Boolean tree-structured null (no epistasis) and alternative (epistasis) distributions on hold-out test data. Finally, our pipeline computes PCS epistasis p-values that probabilisticly quantify improvement in prediction accuracy via bootstrap sampling on the test set. We validate the epiTree pipeline in two case studies using data from the UK Biobank: predicting red hair and multiple sclerosis (MS). In the case of predicting red hair, epiTree recovers known epistatic interactions surrounding MC1R and novel interactions, representing non-linearities not captured by logistic regression models. In the case of predicting MS, a more complex phenotype than red hair, epiTree rankings prioritize novel interactions surrounding HLA-DRB1 , a variant previously associated with MS in several populations. Taken together, these results highlight the potential for epiTree rankings to help reduce the design space for follow up experiments.

59 BASIC BIOLOGICAL SCIENCES↗

Collagen Turnover in Relation to Risk Factors and Hemodynamics in Human Intracranial Aneurysms

Determinants for molecular and structural instability, that is, impending growth or rupture, of intracranial aneurysms (IAs) remain uncertain. To elucidate this, we endeavored to estimate the actual turnover rates of the main molecular constituent in human IA (collagen) on the basis of radiocarbon ( 14 C) birth dating in relation to IA hemodynamics. Collagen turnover rates in excised human IA samples were calculated using mathematical modeling of 14 C birth dating data of collagen in relation to risk factors and histological markers for collagen maturity/turnover in selected IA. Hemodynamics were simulated using image-based computational fluid dynamics. Correlation, logistic regression, and receiver operating characteristic analyses were performed. Collagen turnover rates were estimated in 46 IA (43 patients); computational fluid dynamics could be performed in 20 IA (20 patients). The mean collagen turnover rate (γ) constituted 126% (±1% error) per year. For patients with arterial hypertension, γ was greater than 2600% annually, whereas γ was distinctly lower with 32% (±1% error) per year for patients without risk factors, such as smoking and hypertension. There was a distinct association between histological presence of rather immature collagen in human IA and the presence of modifiable risk factors. Spatial-temporal averaged wall shear stress predicted rapid collagen turnover (odds ratio, 1.6 [95% CI, 1.0–2.7]). Receiver operating characteristic analysis demonstrated a good test accuracy (area under the curve, 0.798 [95% CI, 0.598–0.998]) for average wall shear stress with a threshold ≥4.9 Pa for rapid collagen turnover. Our data indicate that turnover rates and stability of collagen in human IA are strongly associated with the presence of modifiable risk factors and aneurysmal hemodynamics. Finally, these findings underline the importance of strict risk factor modification in patients with unruptured IA. Future should include more detailed risk factor data to establish a more causal understanding of hemodynamics and the rupture risk of individual IA.

60 APPLIED LIFE SCIENCES↗

Comparison of Social Media, Syndromic Surveillance, and Microbiologic Acute Respiratory Infection Data: Observational Study

Internet data can be used to improve infectious disease models. However, the representativeness and individual-level validity of internet-derived measures are largely unexplored as this requires ground truth data for study. This study sought to identify relationships between Web-based behaviors and/or conversation topics and health status using a ground truth, survey-based dataset. This study leveraged a unique dataset of self-reported surveys, microbiological laboratory tests, and social media data from the same individuals toward understanding the validity of individual-level constructs pertaining to influenza-like illness in social media data. Logistic regression models were used to identify illness in Twitter posts using user posting behaviors and topic model features extracted from users’ tweets. Of 396 original study participants, only 81 met the inclusion criteria for this study. Of these participants’ tweets, we identified only two instances that were related to health and occurred within 2 weeks (before or after) of a survey indicating symptoms. It was not possible to predict when participants reported symptoms using features derived from topic models (area under the curve [AUC]=0.51; P=.38), though it was possible using behavior features, albeit with a very small effect size (AUC=0.53; P≤.001). Individual symptoms were also generally not predictable either. The study sample and a random sample from Twitter are predictably different on held-out data (AUC=0.67; P≤.001), meaning that the content posted by people who participated in this study was predictably different from that posted by random Twitter users. Individuals in the random sample and the GoViral sample used Twitter with similar frequencies (similar @ mentions, number of tweets, and number of retweets; AUC=0.50; P=.19). To our knowledge, this is the first instance of an attempt to use a ground truth dataset to validate infectious disease observations in social media data. The lack of signal, the lack of predictability among behaviors or topics, and the demonstrated volunteer bias in the study population are important findings for the large and growing body of disease surveillance using internet-sourced data.

59 BASIC BIOLOGICAL SCIENCES↗

Risk of work injury among adolescent students from single and partnered parent families

Background Parental involvement in keeping their children safe at work has been examined in a handful of studies, with mixed results. Evidence has suggested that non‐work injury risk is higher among children from single‐parent families, but little is known about their risk for work‐related injuries. Methods Five survey cycles of the Canadian Community Health Survey were pooled to create a nationally representative sample of employed 15–19‐year old students (N = 16,620). Multivariable logistic regression estimated the association between family status and work injury. Results Risk of work‐related repetitive strains (OR:1.24, 95%CI: 0.69–2.22) did not differ by family type. However, children of single parents were less likely to sustain a work injury receiving immediate medical care (OR:0.43, 95%CI: 0.19–0.96). Conclusion Despite advantages and disadvantages related to family types, there is no evidence that work‐related injury risk among adolescents from single parent families is greater than that of partnered‐parent families. Am. J. Ind. Med. 60:285–294, 2017. © 2017 Wiley Periodicals, Inc.

Wong, Imelda S.↗

Factors linked to participant attrition in a longitudinal occupational health surveillance program

For occupational medical screening programs focused on long-term health surveillance, participant attrition is a significant barrier to success. Here we investigate demographic, medical history, and clinical data from National Supplemental Screening Program (NSSP) examinees for association with likelihood of return for a second exam (rescreening). A total of 15,733 individuals completed at least one NSSP exam before December 31, 2016; of those, 4832 also completed a second exam on or before December 31, 2019. Stepwise logistic regression models were used to identify variables associated with whether a participant was rescreened in the NSSP. Individuals were less likely to return for rescreening if they had a history of any cancer; cardiovascular problems; diabetes or kidney disease; or if they used insulin. Age at time of first exam and job site category significantly influenced likelihood of return. Workers categorized as “guests” were more likely to return. Participants were less likely to return if they had an abnormal urinalysis, abnormal pulmonary function, pneumoconiosis, aortic atherosclerosis, or hearing loss at their initial exam. Participants who received a chest X-ray at their initial screening were more likely to return. The presence of health problems is strongly linked to screening program attrition. Participants who are older at the time of their initial screening exam are less likely to return. The discovery of several strong demographic, medical, and job associations reveals the importance for medical screening programs to understand and address factors that influence participant retention and, consequently, the effectiveness of long-term health surveillance activities.

60 APPLIED LIFE SCIENCES↗

Adult height, body mass index change, and body shape change in relation to esophageal squamous cell carcinoma risk: A population‐based case‐control study in China

Abstract The relationship between risk of esophageal squamous cell carcinoma (ESCC) and adult height, changes in individual body mass index (BMI) and body shape is not established. We performed a large population‐based case‐control study, which enrolled a total of 1414 ESCC cases and 1989 controls in a high‐incidence area in China. Using face‐to‐face interview with a structured questionnaire, information on participants' heights, weights, and perceived body shapes at 20 years of age was collected. Additionally, data on weight and perceived body shape among the same participants 10 years prior to ascertainment were collected using the same method. Odd ratios (ORs) of ESCC risk in relation to BMI and body shape were estimated using unconditional logistic regression models. The adjusted results indicated that ESCC risk in adults rapidly rose as height increased, plateauing at 170 cm among men and 157 cm among women. Among participants who were underweight, normal weight, or thinner than body shape 4, body weight loss was associated with increased risk of ESCC, and body weight gain was associated with decreased incidence of ESCC (ORs ranging from 0.40 to 0.76). Notably, however, changes in body weight did not significantly affect ESCC risk among participants who were overweight, obese, or larger than body shape 3. Maintaining a fit body shape and a reasonable BMI is advisable and of vital importance to reduce the risk of ESCC, especially in high‐risk areas.

Yang, Xiaorong↗

COVID ‐19 outcomes in patients with cancer: Findings from the University of California health system database

Abstract Background The interaction between cancer diagnoses and COVID‐19 infection and outcomes is unclear. We leveraged a state‐wide, multi‐institutional database to assess cancer‐related risk factors for poor COVID‐19 outcomes. Methods We conducted a retrospective cohort study using the University of California Health COVID Research Dataset, which includes electronic health data of patients tested for severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) at 17 California medical centers. We identified adults tested for SARS‐CoV‐2 from 2/1/2020–12/31/2020 and selected a cohort of patients with cancer. We obtained demographic, clinical, cancer type, and antineoplastic therapy data. The primary outcome was hospitalization within 30d after the first positive SARS‐CoV‐2 test. Secondary outcomes were SARS‐CoV‐2 positivity and severe COVID‐19 (intensive care, mechanical ventilation, or death within 30d after the first positive test). We used multivariable logistic regression to identify cancer‐related factors associated with outcomes. Results We identified 409,462 patients undergoing SARS‐CoV‐2 testing. Of 49,918 patients with cancer, 1781 (3.6%) tested positive. Patients with cancer were less likely to test positive (RR 0.70, 95% CI: 0.67–0.74, p < 0.001). Among the 1781 SARS‐CoV‐2‐positive patients with cancer, BCR/ABL‐negative myeloproliferative neoplasms (RR 2.15, 95% CI: 1.25–3.41, p = 0.007), venetoclax (RR 2.96, 95% CI: 1.14–5.66, p = 0.028), and methotrexate (RR 2.72, 95% CI: 1.10–5.19, p = 0.032) were associated with greater hospitalization risk. Cancer and therapy types were not associated with severe COVID‐19. Conclusions In this large, diverse cohort, cancer was associated with a decreased risk of SARS‐CoV‐2 positivity. Patients with BCR/ABL‐negative myeloproliferative neoplasm or receiving methotrexate or venetoclax may be at increased risk of hospitalization following SARS‐CoV‐2 infection. Mechanistic and comparative studies are needed to validate findings.

60 APPLIED LIFE SCIENCES↗

Predictive analytics of selections of russet potatoes

We explore the application of machine learning algorithms specifically to enhance the selection process of Russet potato (Solanum tuberosum L.) clones in breeding trials by predicting their suitability for advancement. This study addresses the challenge of efficiently identifying high-yield, disease-resistant, and climate-resilient potato varieties that meet processing industry standards. Leveraging manually collected data from trials in the state of Oregon, we investigate the potential of a wide variety of state-of-the-art binary classification models. The dataset includes 1086 clones, with data on 38 attributes recorded for each clone, focusing on yield, size, appearance, and frying characteristics, with several control varieties planted consistently across four Oregon regions from 2013 to 2021. We conduct a comprehensive analysis of the dataset that includes preprocessing, feature engineering, and imputation to address missing values. We focus on several key metrics such as accuracy, F1-score, and Matthews correlation coefficient (MCC) for model evaluation. The top-performing models, namely a feedforward neural network classifier (Neural Net), a histogram-based gradient boosting classifier (HGBC), and a support vector machine classifier (SVM), demonstrate consistent and significant results. To further validate our findings, we conducted a simulation study using the aims, data-generating mechanisms, estimands, methods, and performance measures (ADEMP) framework, simulating different data-generating scenarios to assess model robustness and performance through true positive, true negative, false positive, and false negative distributions, area under the receiver operating characteristic curve (AUC-ROC) and MCC. The simulation results highlight that non-linear models like SVM and HGBC consistently show higher AUC-ROC and MCC than logistic regression, thus outperforming the traditional linear model across various distributions, and emphasizing the importance of model selection and tuning in agricultural trials. Variable selection further enhances model performance and identifies influential features in predicting trial outcomes. The findings emphasize the potential of machine learning in streamlining the selection process for potato varieties, offering benefits such as increased efficiency, substantial cost savings, and judicious resource utilization. Our study contributes insights into precision agriculture and showcases the relevance of advanced technologies for informed decision-making in breeding programs.

60 APPLIED LIFE SCIENCES↗

Predicting fine-scale distributions of peripheral aquatic species in headwater streams

Headwater species and peripheral populations that occupy habitat at the edge of a species range may hold an increased conservation value to managers due to their potential to maximize intraspecies diversity and species' adaptive capabilities in the context of rapid environmental change. The southern Appalachian Mountains are the southern extent of the geographic range of native Salvelinus fontinalis and naturalized Oncorhynchus mykiss and Salmo trutta in eastern North America. In this paper, we predicted distributions of these peripheral, headwater wild trout populations at a fine scale to serve as a planning and management tool for resource managers to maximize resistance and resilience of these populations in the face of anthropogenic stressors. We developed correlative logistic regression models to predict occurrence of brook trout, rainbow trout, and brown trout for every interconfluence stream reach in the study area. A stream network was generated to capture a more consistent representation of headwater streams. Each of the final models had four significant metrics in common: stream order, fragmentation, precipitation, and land cover. Strahler stream order was found to be the most influential variable in two of the three final models and the second most influential variable in the other model. Greater than 70% presence accuracy was achieved for all three models. The underrepresentation of headwater streams in commonly used hydrography datasets is an important consideration that warrants close examination when forecasting headwater species distributions and range estimates. Finally and additionally, it appears that a relative watershed position metric (e.g., stream order) is an important surrogate variable (even when elevation is included) for biotic interactions across the landscape in areas where headwater species distributions are influenced by topographical gradients.

59 BASIC BIOLOGICAL SCIENCES↗

Scavenger removal of bird carcasses at simulated wind turbines: Does carcass type matter?

Abstract Wind energy development can negatively impact bird populations due to bird–turbine collisions. To accurately estimate bird mortality at wind farms, the number of dead birds found under turbines is commonly corrected for carcass removal by scavengers, which is quantified by measuring persistence of experimental carcasses through time. These studies often use domestic birds as surrogates because carcasses of wild birds (e.g., raptors) are difficult to obtain. We assessed scavenger removal of carcasses from five bird species at simulated turbines to determine whether domestic surrogates are scavenged at a different rate than raptors, species of interest for wind turbine mortality. The percentage of carcasses scavenged during 14‐d rounds ranged from 34.6% for American kestrels ( Falco sparverius ) to 65.4% for chickens ( Gallus gallus ), and the percentage of carcasses completely removed ranged from 13.5% for red‐tailed hawks ( Buteo jamaicensis ) to 67.3% for northern bobwhites ( Colinus virginianus ). Carcass type (i.e., species) was the only predictor included in the best‐fit logistic regression model of complete carcass removal, and a survival analysis indicated carcass type influenced elapsed time to scavenging events. Our results suggest the use of surrogate species to quantify carcass removal at wind turbines could lead to inaccurate mortality estimates.

17 WIND ENERGY↗

Variation in lung cancer risk by smoky coal subtype in Xuanwei, China

Lung cancer rates in Xuanwei County have been among the highest in China for both males and females and have been causally associated with exposure to indoor smoky (bituminous) coal emissions that contain very high levels of polycyclic aromatic hydrocarbons. There are numerous coal mines across the County. Although lung cancer risk is strongly associated with the use of smoky coal as a whole, variation in risk by smoky coal subtype has not been characterized as yet. We conducted a population-based case-control study of 498 lung cancer cases and 498 controls, individually matched to case subjects on age and sex to examine risk by coal subtype. Odds ratios (ORs) and 95% confidence intervals (CIs) for coal subtype were calculated by conditional logistic regression, adjusting for potential confounders. Overall, smoky coal use was positively and statistically significantly associated with lung cancer risk, when compared with the use of smokeless coal or wood (OR = 7.7, 95% CI = 4.5-13.3). Furthermore, there was a marked heterogeneity in risk estimates for specific subtypes of smoky coal (test for heterogeneity: p = 5.17 x 10{sup -10}). Estimates were highest for coal of the Laibin (OR = 24.8, 95 % CI = 12.4-49.6) and Longtan (OR = 11.6, 95 % CI = 5.0-27.2) coal types and lower for coal from other subtypes. These findings strongly suggest that in Xuanwei and elsewhere, the carcinogenic potential of coal combustion products can exhibit substantial local variation by specific coal source.

01 COAL, LIGNITE, AND PEAT↗

Nonsteroidal Anti-Inflammatory Drug Prescriptions Are Associated With Increased Stress Fracture Diagnosis in the US Army Population

Stress fractures are common in military personnel and endurance athletes, and nonsteroidal anti-inflammatory drug (NSAID) use is widespread in these populations. NSAIDs inhibit prostaglandin synthesis, which blunts the anabolic response of bone to physical activity and could therefore increase risk of stress fracture. The objective of this study was to determine whether prescribed NSAIDs were associated with stress fracture diagnoses among US Army soldiers. We also aimed to establish whether acetaminophen, an analgesic alternative to NSAIDs, was associated with stress fracture risk. A nested case-control study was conducted using data from the Total Army Injury and Health Outcomes Database from 2002 to 2011 (n = 1,260,168). We identified soldiers with a diagnosis of stress fracture (n = 24,146) and selected 4 controls per case matched on length of military service (n = 96,584). We identified NSAID and acetaminophen prescriptions 180 to 30 days before injury (or match date). We also identified soldiers who participated in basic combat training (BCT), a 10-week period of heightened physical activity at the onset of Army service. Among these individuals, we identified 9088 cases and 36,878 matched controls. Conditional logistic regression was used to calculate incident rate ratios (RR) for stress fracture with adjustment for sex. NSAID prescription was associated with a 2.9-fold increase (RR = 2.9, 95% confidence interval [CI] 2.8–2.9) and acetaminophen prescription with a 2.1-fold increase (RR = 2.1, 95% CI 2.0–2.2) in stress fracture risk within the total Army population. The risk was more than 5-fold greater in soldiers prescribed NSAIDs (RR = 5.3, 95% CI 4.9–5.7) and more than 4-fold greater in soldiers prescribed acetaminophen (RR= 4..4, 95% CI 3.9–4.9) during BCT. Our results reveal an association between NSAID and acetaminophen prescriptions and stress fracture risk, particularly during periods of heightened physical activity. Prospective observational studies and randomized controlled trials are needed to support these findings before clinical recommendations can be made.

60 APPLIED LIFE SCIENCES↗

How Small Can We Go? Evaluating Survival, Tag Retention, and Growth of Juvenile Chinook Salmon Implanted with a New Acoustic Microtag

Abstract Recently, an acoustic microtag, called the Eel–Lamprey Acoustic Tag (ELAT), was developed for use in juvenile American Eel Anguilla rostrata and Pacific Lamprey Entosphenus tridentatus. The ELAT (12.0 × 2.0 mm; 0.088 g in air) has the potential to be used to study other small-bodied fishes. In this pilot study, juvenile Chinook Salmon Oncorhynchus tshawytscha 52–99 mm fork length (FL) were surgically implanted with an ELAT (n = 99) or remained untagged (control; n = 101). Fish were held for 30 d to evaluate survival, tag retention, and growth. Survival did not differ significantly between control (n = 98 or 97.0%) and tagged fish (n = 95 or 95.9%). Logistic regression analysis suggested that survival was influenced by initial FL and tag burden. Mortalities only occurred in tagged fish with initial FLs of ≤65 mm (FL range = 57–65 mm; tag burdens = 3.8–5.9%). However, a tagged fish with an initial FL of 62 mm or an initial tag burden of 4.1% (FL range = 58–64 mm) was predicted to have a 90% chance of survival, and the smallest fish that survived the study was 54 mm FL (1.8 g; tag burden = 4.9%). Tag retention was high (99%), with only one fish expelling an ELAT during the study. Control fish grew slightly larger than the tagged individuals, but differences in final pooled FLs and weight gains were not significant. Our results demonstrate the successful surgical implantation of the ELAT into Chinook Salmon as small as 54 mm FL (1.8 g).

59 BASIC BIOLOGICAL SCIENCES↗

Risk assessment of engineering diseases of embankment–bridge transition section for railway in permafrost regions

Abstract The embankment–bridge transition section (EBTS) is one of the zones where railway diseases occur frequently in permafrost regions. Disease risk assessment of EBTSs can provide guidance for maintenance. In this study, considering the engineering geological conditions, climate characteristics, and embankment structure types along the Qinghai–Tibet Railway (QTR) as well as based on the disease inventory of the QTR from 2010 to 2019, the logistic regression (LR), support vector machine (SVM), and combination‐weight‐based gay relation analysis (GRA) were used for disease risk assessment of the EBTSs along the QTR in permafrost regions. The results indicate that the LR and SVM models have a better capability for EBTS disease prediction than the GRA model, and the SVM model can select more disease samples in relatively larger regions than the LR model. Based on the SVM and LR models, the risk level of EBTSs is divided into four classes: low‐ (29.9%), moderate‐ (39.6%), high‐ (22.1%), and very high (8.4%) risk. Finally, we selected 272 EBTSs in high‐ and very‐high‐risk classes for key observation during the maintenance of the QTR in permafrost regions. This study provides a reference for the risk assessment of railways built in permafrost regions using data‐driven methods.

Zhang, Saize↗

Machine learning-based prediction of enzyme substrate scope: Application to bacterial nitrilases

Predicting the range of substrates accepted by an enzyme from its amino acid sequence is challenging. Although sequenc- and structure-based annotation approaches are often accurate for predicting broad categories of substrate specificity, they generally cannot predict which specific molecules will be accepted as substrates for a given enzyme, particularly within a class of closely related molecules. Combining targeted experimental activity data with structural modeling, ligand docking, and physicochemical properties of proteins and ligands with various machine learning models provides complementary information that can lead to accurate predictions of substrate scope for related enzymes. Here we describe such an approach that can predict the substrate scope of bacterial nitrilases, which catalyze the hydrolysis of nitrile compounds to the corresponding carboxylic acids and ammonia. Each of the four machine learning models (logistic regression, random forest, gradient-boosted decision trees, and support vector machines) performed similarly (average ROC = 0.9, average accuracy = ~82%) for predicting substrate scope for this dataset, although random forest offers some advantages. Finally, this approach is intended to be highly modular with respect to physicochemical property calculations and software used for structural modeling and docking.

59 BASIC BIOLOGICAL SCIENCES↗

Factors affecting stranding of juvenile salmonids by wakes from ship passage in the Lower Columbia River

The effects of deep-draft vessel traffic in confined riverine channels on shorelines and fish are of widespread concern. In the Pacific Northwest of the United States, wakes and subsequent beach run-up from ships transiting the Lower Columbia River have been observed to strand juvenile salmon and other fish. As part of a before-and-after study to assess stranding effects that may be associated with channel deepening, we measured 19 co-variables from observations of 126 vessel passages at three low-slope beaches and used multiple logistic regression to discern the significant factors influencing the frequency of stranding. Subyearling Chinook salmon were 82% of the fish stranded over all sites and seasons. Given a low-slope beach, stranding frequencies for juvenile salmon were significantly related to river location, salmon density in the shallows, a proxy for ship kinetic energy, tidal height, and two interactions. The beach types selected for our study do not include all the beach types along the Lower Columbia River so that the stranding probabilities described here cannot be extrapolated river-wide. A more sophisticated modeling effort, informed by additional field data, is needed to assess salmon losses by stranding for the entire lower river. Such modeling needs to include river-scale factors such as beach type, berms, proximity to navigation channel, and perhaps, proximity to tributaries that act as sources of out-migrating juvenile salmon. At both river and beach scales, no one factor produces stranding; rather interactions among several conditions produce a stranding event and give stranding its episodic nature.

54 ENVIRONMENTAL SCIENCES↗