Radiotherapy treatment-response of cancer patients can vary considerably, even in patients sharing the same diagnosis. Enhancing the degree of treatment personalization might offer a way towards improving curation rates. The recent advancements in the field of deep neural networks provide new directions for the non-invasive extraction of patient-individual biomarkers when applied on diagnostic imaging data. Within this thesis, we explored the potential of image-based deep learning as an enabler for individualized therapy. In a cohort of head and neck cancer patients, we first assessed the suitability of applying convolutional neural networks (CNNs) on pre-treatment computed tomography imaging data for the prediction of loco-regional tumor control in the presence of censored outcomes. We further investigated whether the predictive performance can be improved through the adoption of multitask learning strategies that combine multiple outcome prediction models and a tumor segmentation task, both for CNNs and the recently emerged vision transformer-based network architectures. Subsequently, we applied neural networks on multimodal and longitudinal imaging data collected during the course of radiotherapy and evaluated their potential to further improve outcome models. Finally, in the context of proton-beam radiotherapy of primary brain tumor patients, we applied CNNs for the prediction of the linear energy transfer and examined the feasibility of this approach for estimating treatment-related side-effects considering a variable biological effectiveness of protons.:1. Introduction 2. Theoretical background 3. Convolutional neural networks for outcome prediction 4. Multitask learning and attention-based models for outcome prediction 5. Longitudinal and multimodal models for outcome prediction 6. Prediction of the linear energy transfer in proton-beam radiotherapy / Das Ansprechverhalten von Krebspatienten auf eine Strahlentherapie kann selbst zwischen Patienten mit gleicher Diagnose deutlich variieren. Eine erweiterte Personalisierung der Behandlung stellt einen möglichen Weg dar, um die Heilungschancen zu verbessern. Durch die kürzlich erzielten Fortschritte im Bereich neuronaler Netzwerke bieten sich neue Möglichkeiten zur nicht-invasiven Extraktion patientenspezifischer Merkmale aus diagnostischen Bildgebungsdaten. In dieser Arbeit wurde das Potential tiefer, bildbasierter neuronaler Netzwerke für eine individualisierte Strahlentherapie analysiert. In einer Kohorte von Patienten mit Kopf- und Halstumoren wurde zunächst die Eignung faltender neuronaler Netzwerke (CNNs) zur Vorhersage der lokoregionären Kontrolle im Beisein zensierter Beobachtungen basierend auf prätherapeutischer Computertomographie (CT) evaluiert. Außerdem wurde untersucht, ob sich die Vorhersagequalität von CNNs und den kürzlich vorgestellten vision transformer Architekturen unter Verwendung von Multitask-Lernstrategien, welche mehrere Vorhersagemodelle und eine Segmentierung des Tumors kombinieren, verbessern lässt. Anschließend wurde die Anwendbarkeit neuronaler Netzwerke zur Analyse multimodaler und longitudinaler, während des Behandlungsverlaufes aufgenommener, Bildgebungsdaten aufgezeigt und ihr Potential zur weiteren Verbesserung der Vorhersagen analysiert. Abschließend wurden CNNs im Rahmen der Protonentherapie genutzt, um den linearen Energietransfer bei Patienten mit primären Hirntumoren vorherzusagen. Die Plausibilität dieser Vorhersagen zur Risikoabschätzung behandlungsbedingter Nebenwirkungen wurde unter Berücksichtigung einer variablen relativen biologischen Wirksamkeit von Protonen untersucht.:1. Introduction 2. Theoretical background 3. Convolutional neural networks for outcome prediction 4. Multitask learning and attention-based models for outcome prediction 5. Longitudinal and multimodal models for outcome prediction 6. Prediction of the linear energy transfer in proton-beam radiotherapy

Wir erhoben Daten von 65 Patientinnen, die zwischen 1968 und 2000 erstmalig an einem Granulosa-zelltumor des Ovars erkrankten. Das mediane Follow-up betrug 75 Monate (6-315) Monate. Zur statistischen Auswertung verwende-ten wir die Kaplan-Meier Methode um Überlebensraten zu bestimmen. Sie lagen nach zehn Jahren bei 87% im Stadium I, 75% im Stadium II, 20% im Stadium III und 0% im Sta-dium IV (FIGO). Die Gesamtüber-lebensrate beträgt 76% nach zehn Jahren. Das Tumorstadium, ein postoperativer Residualtumor und der Mitoseindex stellten sich in unserer Studie als wich-tigste prognostische Faktoren dar (p / In a multi-institutional study we analyzed 65 patients who were initially treated for a granulosa cell tumor of the ovary between 1968 and 2000. Median follow-up was 75 months (6-315). We applied the Kaplan-Meier method to estimate survival rates. Survival after ten years was 87% for stage I, 75% for Stage II, 20% for stage III and 0% for stage IV tumor disease. Overall survival was 76% after ten years. Mitotic rate, staging and residual tumor disease were associated with a poor prognosis in our study (p

In survival analysis, proportional hazards model is the most commonly used and the Cox model is the most popular. These models are developed to facilitate statistical analysis frequently encountered in medical research or reliability studies. In analyzing real data sets, checking the validity of the model assumptions is a key component. However, the presence of complicated types of censoring such as double censoring and partly interval-censoring in survival data makes model assessment difficult, and the existing tests for goodness-of-fit do not have direct extension to these complicated types of censored data. In this work, we use empirical likelihood (Owen, 1988) approach to construct goodness-of-fit test and provide estimates for the Cox model with various types of censored data. Specifically, the problems under consideration are the two-sample Cox model and stratified Cox model with right censored data, doubly censored data and partly interval-censored data. Related computational issues are discussed, and some simulation results are presented. The procedures developed in the work are applied to several real data sets with some discussion.

Stacy (1962)首先提出廣義伽瑪分配 (generalized gamma distribution)，此分布被廣泛應用於存活分析 (survival analysis) 以及可靠度 (reliability) 中壽命時間的資料描述。事實上，像是指數分配 (exponential distribution)、韋伯分配 (Weibull distribution) 以及伽瑪分配 (gamma distribution) 都是廣義伽瑪分配的一個特例。 Bologna (1987)提出一個特殊的雙變量廣義伽瑪分配 (bivariate generalized gamma distribution) 可以經由雙變量常態分配 (bivariate normal distribution) 所推得。我們根據他的想法，提出多變量廣義伽瑪分配可以經由多變量常態分配所推得。在過去的研究中，學者們做了許多有關雙變量伽瑪分配。當我們提到雙變量常態分配，由於其分配的型式為唯一的，所以沒人任何人對其分配的型式有疑問。然而，雙變量伽瑪分配卻有很多不同的型式。 在這篇論文中的架構如下。在第二章中，我們介紹並討論雙變量廣義伽瑪分配可以經由雙變量常態分配所推得，接著推導參數估計以及介紹模擬的程序。在第三章中，我們介紹一些對稱以及非對稱的雙變量伽瑪分配，接著拓展到雙變量廣義伽瑪分配，有關參數的估計以及模擬結果也將在此章中討論。在第三章最後，我們建構參數的敏感度分析 (sensitivity analysis)。最後，在第四章中，我們陳述結論以及未來研究方向。 / The generalized gamma distribution was introduced by Stacy (1962). This distribution is useful to describe lifetime data when conducting survival analysis and reliability. In fact, it includes the widely used exponential, Weibull, and gamma distributions as special cases. Bologna (1987) showed that a special bivariate genenralized gamma distribution can be derived from a bivariate normal distribution. Follow his idea, we show that a multivariate generalized gamma distribution can be derived from a multivariate normal distribution. In the past, researchers spend much time in working on a bivariate gamma distribution. When a bivariate normal distribution is mentioned, no one feels puzzled about its form, since it has only one form. However, there are various forms of bivariate gamma distributions. In this paper is as following. In Chapter 2, we introduce and discuss the bivariate generalized gamma distribution, then the multivariate generalized gamma distribution is derived. We also develop parameters estimation and simulation procedure. In Chapter 3, we introduce some symmetrical and asymmetrical bivariate gamma distributions, then they are extended to the bivariate generalized gamma distributions. Problems of parameters estimation and simulation results are also discussed in Chapter 3. Besides, sensitivity analyses of parameters estimation are conducted. Finally, we state conclusion and future work in Chapter 4.

本研究延續Shumway (2001) 從存活分析(Survival Analysis)觀點切入，利用離散型風險模型(Discrete-time Hazard Model)──亦即Shumway 所稱之多期邏輯斯迴歸模型(Multi-period Logistic Regression Model)，建立企業財務危機預警模型。研究選取1986 年至2008 年間718 家上市公司，其中110 家發生財務危機事件，共計6,782 公司/年資料 (firm-year)。有別於Shumway 提出的Log 基期風險型式，本文根據事件發生率圖提出Quadratic 基期風險型式，接著利用4組（或基於會計測量，或基於市場測量）時間相依共變量 (Time-dependent Covariate)建立2 組離散型風險模型(Log 與Quadratic)，並與傳統僅考量單期資料的邏輯斯迴歸模型比較。實證結果顯示，離散型風險模型的解釋變數與破產機率皆符合預期關係，而傳統邏輯斯迴歸模型則有時會出現不符合預期關係的情況；研究亦顯示離散型風險模型預測能力絕大多數情況下優於傳統邏輯斯迴歸模型，在所有模型組合中，以Quadratic 基期風險型式搭配財務變數、市場變數的解釋變數組合而成的離散型風險模型，擁有最佳預測能力。 / Based on the viewpoint of survival analysis from Shumway (2001), the presentthesis utilizes discrete-time hazard model, also called multi-period logistic regression model, to forecast corporate financial distress. From 1986 to 2008, this research chooses 718 listed companies within, which includes 110 failures, as the subjects, summing to 6,782 firm-year data. Being different from Shumway’s log baseline hazard form,we proposed to use quadratic baseline hazard form according to empirical evidence. Then, four groups of time-dependent covariates, which are accounting-based measure or market-based measure, are applied to build two sets of discrete-time hazard model, which is compared with the single-period logistic regression model. The results show that there exists the expected relationship between covariates and predict probability in discrete-time hazard model, while there sometimes lacks it in single-period logistic regression model. The results also show that discrete-time hazard model has better predictive capability than single-period logistic regression model. The model, which combines quadratic baseline hazard form with market and accounting variables, has the best predictive capability among all models.

Approved for public release, distribution is unlimited / The purpose of this thesis is to firstly analyze the determinants on the survival of United States Marine Corps Officers, and secondly, to develop the methodology to optimize the accessions of prior and non-prior enlisted officers. Using data from the Marine Corps Officer Accession Career file (MCCOAC), the Cox Proportional Hazards Model is used to estimate the effects of officer characteristics on their survival as a commissioned officer in the USMC. A Markov model for career transition is combined with fiscal data to determine the optimum number of prior and non-prior enlisted officers under the constraints of force structure and budget. The findings indicate that prior enlisted officers have a better survival rate than their non-prior enlisted counterparts. Additionally, officers who are married, commissioned through MECEP, graduate in the top third of their TBS class, and are assigned to a combat support MOS have a better survival rate than officers who are unmarried, commissioned through USNA, graduate in the middle third of their TBS class, and are assigned to either combat or combat service support MOS. The findings also indicate that the optimum number of prior enlisted officer accessions may be considerably lower than recent trends and may differ across MOS. Based on the findings; it is recommended that prior enlisted officer accession figures be reviewed. / Major, Australian Army

Approved for public release, distribution is unlimited / This study analyzes the relationship between selection to major in the Marine Corps, and the survival of midgrade officers to the promotion point of major, by investigating the effects of billet assignments. Specifically, this study looks at the influence of the percentage of time spent in the Fleet Marine Forces (FMF), the percentage of time spent in primary military occupation (PMOS) billet assignments, and the effect of having served in combat, recruiting, security forces, joint, and drill field duties. Models were formulated using groundwork established in previous promotion, retention, and attrition studies. Assignment variables were then introduced to the models. To account for officers' choice for continued service vice forced attrition, the sample was restricted to officers who had attained five years of service. Probit regression was used to find the influence of career assignments on the probability of selection; Heckman's correction was used to control for self-selection bias; and, Cox proportionalhazard regression was used, utilizing the same assignment factors, to find the influence of assignments on the likelihood of attrition. The findings indicated that FMF and PMOS ratios above 60 percent had a negative effect on promotion and retention. Also indicated was that time spent outside the PMOS, in "B" billets, had a positive effect on retention. In a time of budgetary constraints, this information may provide assistance to personnel planners as an alternative to pecuniary measures used to maintain and shape the force. / Major, United States Marine Corps

購屋與生育同為家庭生命周期重大事件，在高房價低生育率時代，家庭行為的改變，受到政府機關及社會、經濟學者的關注。過去研究指出，購屋及生育行為的連結建立在對家庭的資源與需求，彼此不僅存在資源競爭關係，亦因家庭生命周期的穩定效果，而提高家庭對另一事件的需求。考量家戶對於擁有房子與孩子的次序偏好相異，其家庭資源分配及需求亦不盡相同，故本研究從家庭資源、需求及家庭偏好探討高房價背景下家戶購屋與生育行為的關係；運用中央研究院調查之華人家庭動態資料庫(PSFD)，針對1934~1984年出生之受訪者資料，建立存活模型進行分析。 探討主題有三：一、探討資源排擠效果與生命周期穩定效果對於購屋及生育行為的影響以及影響程度隨時間變化的情形。二、探討購屋對家戶生育行為的影響途徑。三、探討家戶生育事件對購屋年齡影響之世代差異。實證結果，家庭資源及需求對購屋及生育事件的影響為一動態過程。在高房價時期，先生育家戶婚後購屋機率增加，卻有較年長的男性購屋年齡。而先購屋家戶有較年輕的男性購屋年齡，生育機率卻較低。購屋年齡存在世代差異，先生育家戶的購屋年齡隨世代先提前後增加，先購屋家戶的購屋年齡則呈年輕化趨勢。 本研究成果有助於瞭解家庭購屋及生育行為的關係，作為住宅及衛生福利單位於制定獎勵生育、安親托育或住宅福利政策參考。為讓家庭安心培育下一代，建議政策研擬制定應考量家庭偏好差異性、住宅政策及獎勵生育政策關聯性等，並使房價回歸合理及健全安親托育制度，以有效解決高房價、低生育率的社會問題。

<p>Uvod: Sepsa je jedan od vodećih uzroka smrtnosti u jedinicama intenzivnog lečenja i van njih uprkos implementaciji novih dijagnostičkih i terapijskih protokola &scaron;irom sveta. Multiorganska disfunkcija (MODS), koja predstavlja najtežu formu nepovoljnog toka sepse, je u osnovi svojih patofiziolo&scaron;kih de&scaron;avanja obeležena promenama, koje se de&scaron;avaju na nivou kapilara, pre svega u endotelu. Poremećaji koagulacije koji se javljaju kao posledica ovih promena u endotelu su prepoznati kao jedan od dijagnostičkih kriterijuma prema najnovijim smernicama za dijagnostiku i lečenje sepse, međutim njihov značaj u predviđanju toka i ishoda ovog oboljenja jo&scaron; uvek nije precizno definisan. Cilj istraživanja: Odrediti koncentraciju markera endotelne aktivacije, aktivacije koagulacije, aktivnost prirodnih inhibitora koagulacije i funkcionalnost fibrinolize kod obolelih od sepse u odnosu na njihove vrednosti u zdravoj populaciji. Ispitati mogućnost upotrebe markera endotelne disfunkcije i pokazatelja poremećaja mehanizma hemostaze za postavljanje dijagnoze sepse i predikciju pojave komplikacija. Ispitati mogućnost upotrebe markera endotelne disfunkcije i pokazatelja poremećaja mehanizma hemostaze za procenu ishoda kod obolelih od sepse. Materijal i metode: Istraživanje je sprovedeno analitičkom metodom u formi studije preseka, a obuhvatilo je pacijente lečene na Odeljenju anestezije i reanimacije Urgentnog centra Kliničkog centra Vojvodine i na Klinici za infektivne bolesti Kliničkog centra Vojvodine, u Novom Sadu. Istraživanje je sprovođeno tokom 2012. i 2013. godine u trajanju od dve godine. U studiju je bilo uključeno 180 ispitanika od kojih je 150 imalo postavljenu dijagnozu sepse,a 30 ispitanika su činili kontrolnu grupu su klinički i biohemijski zdravih ispitanika, dobrovoljni davaoci krvi. Ispitanici su kategorisani u četiri grupe u odnosu na kliničko stanje i laboratorijske nalaze unutar prvih 24 časa od prijema: bolesnici sa sepsom, te&scaron;kom sepsom, septičkim &scaron;okom i multiorganskom disfunkcijom na prijemu. Nakon kategorizacije ispitanika, izračunati su APACHE II i SOFA numerički pokazatelji procene težine bolesti ispitanika. U roku 24 časa od trenutka postavljanja dijagnoze sepse, iz uzoraka krvi ispitanika, izvr&scaron;ene su predviđene laboratorijske analize u cilju praćenja endotelne aktivacije, aktivacije koagulacije i inhibicije antikoagulantnih mehanizama. U toku 48 časova od prijema, bolesnici koji nisu imali MODS na prijemu su intenzivno praćeni u cilju evidentiranja razvoja multiorganske disfunkcije, dok su bolesnici koji su imali MODS praćeni radi evidentiranja perzistiranja ili eventualne rezolucije MODS-a. Zdravstveno stanje bolesnika je praćeno tokom 28 dana od trenutka uključivanja u studiju i nakon tog perioda je evidentiran ishod lečenja u smislu preživljavanja ili smrtnog ishoda. Statistička analiza je izvr&scaron;ena pomoću statističkog paketa IBM SPSS 20 Statistics. Podaci su predstavljeni tabelarno i grafički, a statistička značajnost određivana je na nivou p&lt; 0,05. Rezultati: Vrednosti biolo&scaron;kih markera endotelne aktivacije i aktivacije koagulacije su statistički značajno povi&scaron;ene kod obolelih od sepse u odnosu na njihove vrednosti u zdravoj populaciji, dok su vrednosti prirodnih inhibitora koagulacije statistički značajno snižene kod obolelih od sepse u odnosu na njihove vrednosti u zdravoj populaciji. Vrednosti APTT-a, PT-a, D-dimera, fibrinogena, prirodnih inhibitora koagulacije i markera endotelne aktivacije (endokan i vWF antigena i aktivnosti) imaju značajan i veoma visok dijagnostički potencijal. Vrednosti biomarkera endotelne disfunkcije i pokazatelja poremećaja hemostaznog mehanizma su značajni prediktori komplikacija kod bolesnika sa sepsom. APTT, PT, D-dimer, broj trombocita, vrednosti priorodnih inhibitora koagulacije, trombomodulina, endokana i ETP-a su jednako validni u inicijalnoj proceni toka kliničke slike sepse kao i prediktivni APACHE II i SOFA skorovi. Koncentracija trombomodulina, D-dimera, ETP-a i PC su dobri prediktori nastanka MODS-a u prvih 48 časova u toku sepse. Endokan, PT, APTT, koncentracija fibrinogena, prirodnih inhibitora koagulacije i vrednosti ETP-a su značajni u predikciji mortaliteta kod bolesnika sa sepsom. Zaključci: Ukoliko bi pokazatelji aktivacije endotela i mehanizma hemostaze bili inkorporirani u određeni sistem skorovanja u cilju procene težine bolesti u smislu ishoda kod bolesnika sa sepsom, to bi moglo doneti doprinos boljoj klasifikaciji bolesnika, te primeni pravovremene i adekvatne terapije u cilju postizanja pozitivnog ishoda kod bolesnika sa sepsom. Prilikom interpretacije pokazatelja inflamacije i koagulacije neophodno je steći uvid u celokupnu sliku pro-i antikoagulantnih de&scaron;avanja koja se odvijaju tokom sepse, odnosno adekvatno proceniti pravac toka disbalansa mehanizma hemostaze da bi se eventualnim terapijskim merama mogao postići pozitivan učinak.</p> / <p>Introduction: Sepsis is one of the main causes of death in intensive care units and other hospital wards in spite of implementation of new sepsis treatment guidelines in everyday hospital practice worldwide. Changes that occur in the microvasculature, affecting primarily endothelial cell, are the basis of the pathophysiology of multiorgan dysfunction (MODS) in sepsis. Coagulation abnormalities which occur as a consequence of endothelial changes are recognized as diagnostic criteria for sepsis, but significance of these changes in the outcome prognosis and prediction of the course of sepsis is still not accurately defined. Aims: Evaluation of hemostasis related parameters and endothelial activation biomarkers values in patients with sepsis and healthy volunteers. Determination whether the levels of hemostasis-related parameters and biomarkers of endothelial activation have diagnostic significance and are they associated with MODS development and persistence in the first 48 hours of hospitalization and 28-day mortality in patients with sepsis. Material and methods: This is cross-sectional study conducted in 2012 and 2013 in the Department of Anesthesia and Reanimation at the Emergency Center of the Clinical Center of Vojvodina and in the Clinic of Infectious Disease at the Clinical Center of Vojvodina. 150 patients who fulfilled criteria for diagnosis of sepsis were included in the study. Patients were divided into 4 groups: sepsis, severe sepsis, septic shock and MODS. 30 healthy volunteers, blood donors were the control group. After the categorization of patients, during the first 24 hours of hospitalization, predictive APACHE II and SOFA scores were calculated. Hemostasis related parameters and endothelial activation biomarkers concentrations were determined within the first 24 hours of the onset of the disease. To assess the development of complication of the disease, patients were monitored for 48 hours for MODS development and persistence or resolution and for 28 days from the onset of sepsis for outcome assessment. Data were analyzed using SPSS 20.0 software and are presented in tables and graphs, statistical significance was set at p&lt; 0,05. Results: Biomarkers of endothelial and coagulation activation are significantly higher in patients with sepsis in comparison to their values in healthy volunteers, while concentrations of natural anticoagulants are significantly lower in patients with sepsis than in healthy volunteers. APTT, PT, D-dimer, fibrinogen, natural anticoagulants and biomarkers od endothelial activation (endocan and vWF antigen and activity) have diagnostic significance in patients with sepsis. Hemostasis related parameters and endothelial activation biomarkers are good prognostic factors for complication development in patients with sepsis. APTT, PT, D-dimer, platelet count, natural anticoagulants, thrombomodulin, endocan and ETP are equally valuable in early prediction of sepsis development as APACHE II and SOFA scores. Thrombomodulin, D-dimer, ETP and PC are good predictors of MODS development during the first 48 hours from sepsis onset. Endocan, PT, APTT, fibrinogen concentration, values of natural anticoagulants and ETP values are significant in 28-day mortality prediction in patients with sepsis. Conclusion: A combination of markers of endothelial dysfunction with widely used ICU scores and organ failure assessment could contribute to an early recognition of complication development and consequent death in patients with sepsis. It is necessary to obtain the full insight in pro-and anticoagulant dynamic evaluation while interpreting coagulation and inflammation processes in sepsis development, in order to accurately lead early resuscitation therapy.</p>

The evolution of extra-group paternity (EGP) is a contentious issue in evolutionary biology. This thesis examines the factors and adaptive benefits driving EGP in a high-density, group-living population of European badgers (Meles meles). To improve power to assign parentage, I isolated and characterised 21 new polymorphic microsatellite markers. I genotyped 83% of 1410 badger trapped 1987‒2010 using 35 autosomal microsatellite markers. Maternity and paternity were assigned at 80% confidence ca. 82% of individuals. 48% of paternities were extra-group, where 85% were attributable to neighbouring-group males and EGP was detected in 47% of litters; thus badger social group do not correspond with a breeding unit. I tested whether indirect genetic benefits explain these high EGP rates. (1) ‘Good-gene-as-heterozygosity Hypothesis’: Paternal heterozygosity, but not maternal or an individual’s own heterozygosity, associated positively with first-year survival probability. Under benign environmental conditions, cubs fathered by more heterozygous males had a higher first year survival probability. Despite this correlation, the EGP rate per litter correlated with neither average nor maximum within-group heterozygosity of candidate fathers. (2) Fitness benefit Hypothesis: Extra-group offspring (EGO) had lower first-year survival probability and lived 1.3 years less than within-group offspring (WGO). Female WGO produced more litters and offspring over their lifetime than female EGO, whereas male EGO produced more offspring than male WGO. (3) Inbreeding avoidance hypothesis: The EGP rate within a litter increased with greater average pair-wise relatedness between mothers and within-group candidate fathers. No inbreeding depression on first-year survival probability was detected, but small sample sizes limited statistical power. Socio-ecologically, at the litter level, EGP correlated negatively with the number of within-group candidate fathers, and positively with neighbouring-group candidate fathers. In conclusion, EGP in badgers may reduce inbreeding and be maintained in the population through a sex-specific antagonistic selection and indirect genetic benefits may occur when the total fitness benefits of producing extra-group sons outweigh the costs of producing extra-group daughters. These indirect genetic benefits only partially explain the evolution of promiscuity in European badgers, highlighting that evolutionary factors underlying promiscuity remain unclear.