ADMANI: Annotated Digital Mammograms and Associated Non-Image Datasets.

Supplemental material is available for this article. Keywords: Mammography, Screening, Convolutional Neural Network (CNN) Published under a CC BY 4.0 license. See also the commentary by Cadrin-Chênevert in this issue.

The evolution of two transmissible cancers in Tasmanian devils.

Tasmanian devils have spawned two transmissible cancer lineages, named devil facial tumor 1 (DFT1) and devil facial tumor 2 (DFT2). We investigated the genetic diversity and evolution of these clones by analyzing 78 DFT1 and 41 DFT2 genomes relative to a newly assembled, chromosome-level reference. Time-resolved phylogenetic trees reveal that DFT1 first emerged in 1986 (1982 to 1989) and DFT2 in 2011 (2009 to 2012). Subclone analysis documents transmission of heterogeneous cell populations. DFT2 has faster mutation rates than DFT1 across all variant classes, including substitutions, indels, rearrangements, transposable element insertions, and copy number alterations, and we identify a hypermutated DFT1 lineage with defective DNA mismatch repair. Several loci show plausible evidence of positive selection in DFT1 or DFT2, including loss of chromosome Y and inactivation of MGA, but none are common to both cancers. This study reveals the parallel long-term evolution of two transmissible cancers inhabiting a common niche in Tasmanian devils.

Chest-Based Wearables and Individualized Distributions for Assessing Postural Sway in Persons With Multiple Sclerosis.

Typical assessments of balance impairment are subjective or require data from cumbersome and expensive force platforms. Researchers have utilized lower back (sacrum) accelerometers to enable more accessible, objective measurement of postural sway for use in balance assessment. However, new sensor patches are broadly being deployed on the chest for cardiac monitoring, opening a need to determine if measurements from these devices can similarly inform balance assessment. Our aim in this work is to validate postural sway measurements from a chest accelerometer. To establish concurrent validity, we considered data from 16 persons with multiple sclerosis (PwMS) asked to stand on a force platform while also wearing sensor patches on the sacrum and chest. We found five of 15 postural sway features derived from the chest and sacrum were significantly correlated with force platform-derived features, which is in line with prior sacrum-derived findings. Clinical significance was established using a sample of 39 PwMS who performed eyes-open, eyes-closed, and tandem standing tasks. This cohort was stratified by fall status and completed several patient-reported measures (PRM) of balance and mobility impairment. We also compared sway features derived from a single 30-second period to those derived from a one-minute period with a sliding window to create individualized distributions of each postural sway feature (ID method). We find traditional computation of sway features from the chest is sensitive to changes in PRMs and task differences. Distribution characteristics from the ID method establish additional relationships with PRMs, detect differences in more tasks, and distinguish between fall status groups. Overall, the chest was found to be a valid location to monitor postural sway and we recommend utilizing the ID method over single-observation analyses.

How Much Data Is Enough? A Reliable Methodology to Examine Long-Term Wearable Data Acquisition in Gait and Postural Sway.

Wearable sensors facilitate the evaluation of gait and balance impairment in the free-living environment, often with observation periods spanning weeks, months, and even years. Data supporting the minimal duration of sensor wear, which is necessary to capture representative variability in impairment measures, are needed to balance patient burden, data quality, and study cost. Prior investigations have examined the duration required for resolving a variety of movement variables (e.g., gait speed, sit-to-stand tests), but these studies use differing methodologies and have only examined a small subset of potential measures of gait and balance impairment. Notably, postural sway measures have not yet been considered in these analyses. Here, we propose a three-level framework for examining this problem. Difference testing and intra-class correlations (ICC) are used to examine the agreement in features computed from potential wear durations (levels one and two). The association between features and established patient reported outcomes at each wear duration is also considered (level three) for determining the necessary wear duration. Utilizing wearable accelerometer data continuously collected from 22 persons with multiple sclerosis (PwMS) for 6 weeks, this framework suggests that 2 to 3 days of monitoring may be sufficient to capture most of the variability in gait and sway; however, longer periods (e.g., 3 to 6 days) may be needed to establish strong correlations to patient-reported clinical measures. Regression analysis indicates that the required wear duration depends on both the observation frequency and variability of the measure being considered. This approach provides a framework for evaluating wear duration as one aspect of the comprehensive assessment, which is necessary to ensure that wearable sensor-based methods for capturing gait and balance impairment in the free-living environment are fit for purpose.

Restoring faith in conservation action: Maintaining wild genetic diversity through the Tasmanian devil insurance program.

Conservation breeding programs aim to maintain 90% wild genetic diversity, but rarely assess functional diversity. Here, we compare both genome-wide and functional diversity (in over 500 genes) of Tasmanian devils (Sarcophilus harrisii) within the insurance metapopulation and across the species' range (64,519 km2). Populations have declined by 80% since 1996 due to a contagious cancer, devil facial tumor disease (DFTD). However, predicted local extinctions have not occurred. Recent suggestions of selection for "resistance" alleles in the wild precipitated concerns that insurance population devils may be unsuitable for translocations. Using 830 wild samples collected at 31 locations between 2012 and 2021, and 553 insurance metapopulation devils, we show that the insurance metapopulation is representative of current wild genetic diversity. Allele frequencies at DFTD-associated loci were not substantially different between captive and wild devils. Methods presented here are valuable for others investigating evolutionary potential in threatened species, particularly ones under significant selective pressures.

The Sit-to-Stand Transition as a Biomarker for Impairment: Comparison of Instrumented 30-Second Chair Stand Test and Daily Life Transitions in Multiple Sclerosis.

Falls and mobility deficits are common in people with multiple sclerosis (PwMS) across all levels of clinical disability. However, functional mobility observed in supervised settings may not reflect daily life which may impact assessments of fall risk and impairment in the clinic. To investigate this further, we compared the utility of sensor-based performance metrics from sit-stand transitions during daily life and a structured task to inform fall risk and impairment in PwMS. Thirty-seven PwMS instrumented with wearable sensors (thigh and chest) completed supervised 30-second chair stand tests (30CST) and underwent two days of instrumented daily life monitoring. Performance metrics were computed for sit-stand transitions during daily life and 30CSTs. EDSS sub scores and fall history were used to dichotomize participants into groups: pyramidal/no pyramidal impairment, sensory/no sensory impairment and high/low fall risk. The ability of performance metrics to discriminate between groups was assessed using the area under the curve (AUC). The feature that best discriminated between high and low fall risk was a chest acceleration measurement from the supervised instrumented 30CST (AUC = 0.89). Only chest features indicated sensory impairment, however the task was different between supervised and daily life. The metric that best discriminated pyramidal impairment was a chest-derived feature (AUC = 0.89) from supervised 30CSTs. The highest AUC from daily life was observed in faller classification with the average sit-stand time (0.81). While characterizing sit-stand performance during daily life may yield insights into fall risk and may be performed without a clinic visit, there remains value to conducting supervised functional assessments to provide the best classification performance between the investigated impairments in this sample.

Quality of life and functional status of patients treated with venovenous extracorporeal membrane oxygenation at 6 months.

PURPOSE: Quality of life (QoL) outcomes of patients treated with extracorporeal membrane oxygenation (ECMO) for acute respiratory distress syndrome (ARDS) have been conflicting. This study reports on QoL outcomes for a broad group of ARDS patients managed with up-to-date treatment modalities. METHODS: We prospectively recruited patients at a quaternary hospital in the United Kingdom from 2013 to 2015 who were treated with ECMO for ARDS. We evaluated their pulmonary function and QoL at 6-months after admission using three QoL instruments: EuroQoL 5D (EQ-5), HADS, and PTSS-14. RESULTS: Forty-three patients included in the analysis had near-normal pulmonary function at 6 months. HADS showed moderate-to-severe anxiety and depression in 32% and 11% of patients, respectively. PTSS-14 showed 29% had signs of post-traumatic stress disorder. EQ-5D showed that 67% of patients had difficulty returning to usual activities, 74% suffered some pain, none reported severe problems and 77% were able to return to work. No clinical or demographic variables were associated with poor 6-month QoL. CONCLUSIONS: Patients with ARDS treated with ECMO generally had good QoL outcomes, similar to outcomes reported for patients managed without ECMO. With respect to QoL, VV-EMCO represents a valid treatment modality for patients with refractory ARDS.

Evolution and lineage dynamics of a transmissible cancer in Tasmanian devils.

Devil facial tumour 1 (DFT1) is a transmissible cancer clone endangering the Tasmanian devil. The expansion of DFT1 across Tasmania has been documented, but little is known of its evolutionary history. We analysed genomes of 648 DFT1 tumours collected throughout the disease range between 2003 and 2018. DFT1 diverged early into five clades, three spreading widely and two failing to persist. One clade has replaced others at several sites, and rates of DFT1 coinfection are high. DFT1 gradually accumulates copy number variants (CNVs), and its telomere lengths are short but constant. Recurrent CNVs reveal genes under positive selection, sites of genome instability, and repeated loss of a small derived chromosome. Cultured DFT1 cell lines have increased CNV frequency and undergo highly reproducible convergent evolution. Overall, DFT1 is a remarkably stable lineage whose genome illustrates how cancer cells adapt to diverse environments and persist in a parasitic niche.

A heat-shock inducible system for flexible gene expression in cereals.

BACKGROUND: Functional characterisation of genes using transgenic methods is increasingly common in cereal crops. Yet standard methods of gene over-expression can lead to undesirable developmental phenotypes, or even embryo lethality, due to ectopic gene expression. Inducible expression systems allow the study of such genes by preventing their expression until treatment with the specific inducer. When combined with the Cre-Lox recombination system, inducible promoters can be used to initiate constitutive expression of a gene of interest. Yet while these systems are well established in dicot model plants, like Arabidopsis thaliana, they have not yet been implemented in grasses. RESULTS: Here we present an irreversible heat-shock inducible system developed using Golden Gate-compatible components which utilises Cre recombinase to drive constitutive gene expression in barley and wheat. We show that a heat shock treatment of 38 °C is sufficient to activate the construct and drive expression of the gene of interest. Modulating the duration of heat shock controls the density of induced cells. Short durations of heat shock cause activation of the construct in isolated single cells, while longer durations lead to global construct activation. The system can be successfully activated in multiple tissues and at multiple developmental stages and shows no activation at standard growth temperatures (~ 20 °C). CONCLUSIONS: This system provides an adaptable framework for use in gene functional characterisation in cereal crops. The developed vectors can be easily adapted for specific genes of interest within the Golden Gate cloning system. By using an environmental signal to induce activation of the construct, the system avoids pitfalls associated with consistent and complete application of chemical inducers. As with any inducible system, care must be taken to ensure that the expected construct activation has indeed taken place.

Inbreeding depression in one of the last DFTD-free wild populations of Tasmanian devils.

BACKGROUND: Vulnerable species experiencing inbreeding depression are prone to localised extinctions because of their reduced fitness. For Tasmanian devils, the rapid spread of devil facial tumour disease (DFTD) has led to population declines and fragmentation across the species' range. Here we show that one of the few remaining DFTD-free populations of Tasmanian devils is experiencing inbreeding depression. Moreover, this population has experienced a significant reduction in reproductive success over recent years. METHODS: We used 32 microsatellite loci to examine changes in genetic diversity and inbreeding in the wild population at Woolnorth, alongside field data on breeding success from females to test for inbreeding depression. RESULTS: Wefound that maternal internal relatedness has a negative impact on litter sizes. The results of this study imply that this population may be entering an extinction vortex and that to protect the population genetic rescue should be considered. This study provides conservation managers with useful information for managing wild devils and provides support for the "Wild Devil Recovery Program", which is currently augmenting small, isolated populations.

An oral bait vaccination approach for the Tasmanian devil facial tumor diseases.

Introduction: The Tasmanian devil (Sarcophilus harrisii) is the largest extant carnivorous marsupial. Since 1996, its population has declined by 77% primarily due to a clonal transmissible tumor, known as devil facial tumor (DFT1) disease. In 2014, a second transmissible devil facial tumor (DFT2) was discovered. DFT1 and DFT2 are nearly 100% fatal.Areas covered: We review DFT control approaches and propose a rabies-style oral bait vaccine (OBV) platform for DFTs. This approach has an extensive safety record and was a primary tool in large-scale rabies virus elimination from wild carnivores across diverse landscapes. Like rabies virus, DFTs are transmitted by oral contact, so immunizing the oral cavity and stimulating resident memory cells could be advantageous. Additionally, exposing infected devils that already have tumors to OBVs could serve as an oncolytic virus immunotherapy. The primary challenges may be identifying appropriate DFT-specific antigens and optimization of field delivery methods.Expert opinion: DFT2 is currently found on a peninsula in southern Tasmania, so an OBV that could eliminate DFT2 should be the priority for this vaccine approach. Translation of an OBV approach to control DFTs will be challenging, but the approach is feasible for combatting ongoing and future disease threats.

Tracing the rise of malignant cell lines: Distribution, epidemiology and evolutionary interactions of two transmissible cancers in Tasmanian devils.

Emerging infectious diseases are rising globally and understanding host-pathogen interactions during the initial stages of disease emergence is essential for assessing potential evolutionary dynamics and designing novel management strategies. Tasmanian devils (Sarcophilus harrisii) are endangered due to a transmissible cancer-devil facial tumour disease (DFTD)-that since its emergence in the 1990s, has affected most populations throughout Tasmania. Recent studies suggest that devils are adapting to the DFTD epidemic and that disease-induced extinction is unlikely. However, in 2014, a second and independently evolved transmissible cancer-devil facial tumour 2 (DFT2)-was discovered at the d'Entrecasteaux peninsula, in south-east Tasmania, suggesting that the species is prone to transmissible cancers. To date, there is little information about the distribution, epidemiology and effects of DFT2 and its interaction with DFTD. Here, we use data from monitoring surveys and roadkills found within and adjacent to the d'Entrecasteaux peninsula to determine the distribution of both cancers and to compare their epidemiological patterns. Since 2012, a total of 51 DFTD tumours have been confirmed among 26 individuals inside the peninsula and its surroundings, while 40 DFT2 tumours have been confirmed among 23 individuals, and two individuals co-infected with both tumours. All devils with DFT2 were found within the d'Entrecasteaux peninsula, suggesting that this new transmissible cancer is geographically confined to this area. We found significant differences in tumour bodily location in DFTD and DFT2, with non-facial tumours more commonly found in DFT2. There was a significant sex bias in DFT2, with most cases reported in males, suggesting that since DFT2 originated from a male host, females might be less susceptible to this cancer. We discuss the implications of our results for understanding the epidemiological and evolutionary interactions of these two contemporary transmissible cancers and evaluating the effectiveness of potential management strategies.

The prevalence and utility of screening for urinary tract infection at the time of presumed multiple sclerosis relapse.

BACKGROUND: Methods of screening for infections at the time of suspected relapse in people with multiple sclerosis (MS) vary across physicians. People with multiple sclerosis (MS) are at an increased risk of urinary tract infection (UTI). Data evaluating the utility of screening for potential UTI at the time of suspected relapse and whether there are key subgroups of patients in which screening would be most effective are sparse. OBJECTIVES: To evaluate demographic and clinical predictors of UTI in the context of a suspected acute relapse in (1) a retrospective hospital admission cohort and (2) a prospectively-enrolled, ambulatory care-based cohort, and to determine an approximate number needed to screen to detect one UTI in both healthcare settings. METHODS: For the hospital admissions cohort, we included individuals with a known or new diagnosis of MS or clinically isolated syndrome who were admitted at least once to the Johns Hopkins Neurology Inpatient Service (March 2012 to December 2014). We considered those screened via urinalysis. Possible UTI was defined as leukocyte esterase OR nitrite positive. For the ambulatory population, we enrolled a cohort of RRMS patients aged 18-65 who were suspected of suffering from an acute MS relapse who either called or came into clinic. Participants were screened via urinalysis; possible UTI was similarly defined. Participants also completed questionnaires (disability, history of Uhthoff's-type phenomenon, recent sexual intercourse, and new urologic symptoms). For both cohorts, we calculated an approximate number needed to screen, and tested if demographic and patient characteristics were associated with possible UTI using logistic regression models. RESULTS: For the hospital admissions cohort, we included 158 individuals; 48 (30.4%) were identified as possibly having a UTI. For possible UTI, the approximate number needed to screen in order to detect 1 possible UTI is 3 (95% CI: 2, 6). Female sex was the only factor associated with increased odds of UTI (odds ratio [OR]: 3.90; 95% CI: 1.59-9.61; p = 0.003). For the ambulatory cohort, we included 50 participants; 10 (20.0%) with possible UTI. The approximate number needed to screen in order to detect 1 possible UTI was 5 (95% CI: 3, 11) in this cohort. Foul-smelling urine was positively associated with UTI (OR: 5.36; 95% CI: 1.10, 26.17; p = 0.04); no men had a possible UTI in this cohort, so we could not estimate odds ratios associated with sex. CONCLUSION: UTIs at the time of a suspected MS relapse are relatively uncommon. Female sex is a strong risk factor for UTI in people with MS; foul-smelling urine is a potential predictor of UTI in people with MS. Larger studies are needed to comprehensively evaluate the utility of screening and risk factors for UTI at the time of suspected MS relapse.

Systems Biology Approach Pinpoints Minimum Requirements for Auxin Distribution during Fruit Opening.

The phytohormone auxin is implied in steering various developmental decisions during plant morphogenesis in a concentration-dependent manner. Auxin maxima have been shown to maintain meristematic activity, for example, of the root apical meristem, and position new sites of outgrowth, such as during lateral root initiation and phyllotaxis. More recently, it has been demonstrated that sites of auxin minima also provide positional information. In the developing Arabidopsis fruit, auxin minima are required for correct differentiation of the valve margin. It remains unclear, however, how this auxin minimum is generated and maintained. Here, we employ a systems biology approach to model auxin transport based on experimental observations. This allows us to determine the minimal requirements for its establishment. Our simulations reveal that two alternative processes-which we coin "flux-barrier" and "flux-passage"-are both able to generate an auxin minimum, but under different parameter settings. Both models are in principle able to yield similar auxin profiles but present qualitatively distinct patterns of auxin flux. The models were tested by tissue-specific inducible ablation, revealing that the auxin minimum in the fruit is most likely generated by a flux-passage process. Model predictions were further supported through 3D PIN localization imaging and implementing experimentally observed transporter localization. Through such an experimental-modeling cycle, we predict how the auxin minimum gradually matures during fruit development to ensure timely fruit opening and seed dispersal.

Tasmanian devils with contagious cancer exhibit a constricted T-cell repertoire diversity.

The Tasmanian devil (Sarcophilus harrisii) is threatened by a contagious cancer, known as Devil Facial Tumour Disease (DFTD). A highly diverse T-cell receptor (TCR) repertoire is crucial for successful host defence against cancers. By investigating TCR beta chain diversity in devils of different ages, we show that the T-cell repertoire in devils constricts in their second year of life, which may explain the higher DFTD prevalence in older devils. Unexpectedly, we also observed a pronounced decline in TCR diversity and T cell clonal expansion in devils after DFTD infection. These findings overturned the previous assumption that DFTD did not directly impact host immunity.

Complex problems need detailed solutions: Harnessing multiple data types to inform genetic management in the wild.

For bottlenecked populations of threatened species, supplementation often leads to improved population metrics (genetic rescue), provided that guidelines can be followed to avoid negative outcomes. In cases where no "ideal" source populations exist, or there are other complicating factors such as prevailing disease, the benefit of supplementation becomes uncertain. Bringing multiple data and analysis types together to plan genetic management activities can help. Here, we consider three populations of Tasmanian devil, Sarcophilus harrisii, as candidates for genetic rescue. Since 1996, devil populations have been severely impacted by devil facial tumour disease (DFTD), causing significant population decline and fragmentation. Like many threatened species, the key threatening process for devils cannot currently be fully mitigated, so species management requires a multifaceted approach. We examined diversity of 31 putatively neutral and 11 MHC-linked microsatellite loci of three remnant wild devil populations (one sampled at two time-points), alongside computational diversity projections, parameterized by field data from DFTD-present and DFTD-absent sites. Results showed that populations had low diversity, connectivity was poor, and diversity has likely decreased over the last decade. Stochastic simulations projected further diversity losses. For a given population size, the effects of DFTD on population demography (including earlier age at death and increased female productivity) did not impact diversity retention, which was largely driven by final population size. Population sizes ≥500 (depending on the number of founders) were necessary for maintaining diversity in otherwise unmanaged populations, even if DFTD is present. Models indicated that smaller populations could maintain diversity with ongoing immigration. Taken together, our results illustrate how multiple analysis types can be combined to address complex population genetic challenges.

Looking like the locals - gut microbiome changes post-release in an endangered species.

BACKGROUND: Captivity presents extreme lifestyle changes relative to the wild, and evidence of microbiome dysbiosis in captive animals is growing. The gut microbiome plays a crucial role in host health. Whilst captive breeding and subsequent reintroduction to the wild is important for conservation, such efforts often have limited success. Post-release monitoring is essential for assessing translocation success, but changes to the microbiome of released individuals are poorly understood. The Tasmanian devil was previously shown to exhibit loss of microbiome diversity as a result of intense captive management. This current study examines changes in the devil gut microbiome in response to translocation and aims to determine if perturbations from captivity are permanent or reversible. METHODS: Using 16S rRNA amplicon sequencing, we conducted temporal monitoring of the gut microbiome of released devils during two translocation events, captive-to-wild and wild-to-wild. To investigate whether the microbiome of the released devils changed following translocation, we characterized their microbiome at multiple time points during the translocation process over the course of 6-12 months and compared them to the microbiome of wild incumbent devils (resident wild-born devils at the respective release sites). RESULTS: We showed that the pre-release microbiome was significantly different to the microbiome of wild incumbent animals, but that the microbiomes of animals post-release (as early as 3 to 4 weeks post-release) were similar to wild incumbents. The gut microbiome of released animals showed significant compositional shifts toward the wild incumbent microbiome of both translocation events. CONCLUSION: Our results suggest that the devil gut microbiome is dynamic and that loss of microbiome diversity in captivity can be restored following release to the wild. We recommend the broader application of microbiome monitoring in wildlife translocation programs to assess the impacts of translocation on animal microbiomes.

Tasman-PCR: a genetic diagnostic assay for Tasmanian devil facial tumour diseases.

Tasmanian devils have spawned two transmissible cancer clones, known as devil facial tumour 1 (DFT1) and devil facial tumour 2 (DFT2). DFT1 and DFT2 are transmitted between animals by the transfer of allogeneic contagious cancer cells by biting, and both cause facial tumours. DFT1 and DFT2 tumours are grossly indistinguishable, but can be differentiated using histopathology, cytogenetics or genotyping of polymorphic markers. However, standard diagnostic methods require specialist skills and equipment and entail long processing times. Here, we describe Tasman-PCR: a simple polymerase chain reaction (PCR)-based diagnostic assay that identifies and distinguishes DFT1 and DFT2 by amplification of DNA spanning tumour-specific interchromosomal translocations. We demonstrate the high sensitivity and specificity of this assay by testing DNA from 546 tumours and 804 normal devils. A temporal-spatial screen confirmed the reported geographic ranges of DFT1 and DFT2 and did not provide evidence of additional DFT clones. DFT2 affects disproportionately more males than females, and devils can be co-infected with DFT1 and DFT2. Overall, we present a PCR-based assay that delivers rapid, accurate and high-throughput diagnosis of DFT1 and DFT2. This tool provides an additional resource for devil disease management and may assist with ongoing conservation efforts.

Spatiotemporal coordination of cell division and growth during organ morphogenesis.

A developing plant organ exhibits complex spatiotemporal patterns of growth, cell division, cell size, cell shape, and organ shape. Explaining these patterns presents a challenge because of their dynamics and cross-correlations, which can make it difficult to disentangle causes from effects. To address these problems, we used live imaging to determine the spatiotemporal patterns of leaf growth and division in different genetic and tissue contexts. In the simplifying background of the speechless (spch) mutant, which lacks stomatal lineages, the epidermal cell layer exhibits defined patterns of division, cell size, cell shape, and growth along the proximodistal and mediolateral axes. The patterns and correlations are distinctive from those observed in the connected subepidermal layer and also different from the epidermal layer of wild type. Through computational modelling we show that the results can be accounted for by a dual control model in which spatiotemporal control operates on both growth and cell division, with cross-connections between them. The interactions between resulting growth and division patterns lead to a dynamic distributions of cell sizes and shapes within a deforming leaf. By modulating parameters of the model, we illustrate how phenotypes with correlated changes in cell size, cell number, and organ size may be generated. The model thus provides an integrated view of growth and division that can act as a framework for further experimental study.

Large-effect loci affect survival in Tasmanian devils (Sarcophilus harrisii) infected with a transmissible cancer.

Identifying the genetic architecture of complex phenotypes is a central goal of modern biology, particularly for disease-related traits. Genome-wide association methods are a classical approach for identifying the genomic basis of variation in disease phenotypes, but such analyses are particularly challenging in natural populations due to sample size difficulties. Extensive mark-recapture data, strong linkage disequilibrium and a lethal transmissible cancer make the Tasmanian devil (Sarcophilus harrisii) an ideal model for such an association study. We used a RAD-capture approach to genotype 624 devils at ~16,000 loci and then used association analyses to assess the heritability of three cancer-related phenotypes: infection case-control (where cases were infected devils and controls were devils that were never infected), age of first infection and survival following infection. The SNP array explained much of the phenotypic variance for female survival (>80%) and female case-control (>61%). We found that a few large-effect SNPs explained much of the variance for female survival (~5 SNPs explained >61% of the total variance), whereas more SNPs (~56) of smaller effect explained less of the variance for female case-control (~23% of the total variance). By contrast, these same SNPs did not account for a significant proportion of phenotypic variance in males, suggesting that the genetic bases of these traits and/or selection differ across sexes. Loci involved with cell adhesion and cell-cycle regulation underlay trait variation, suggesting that the devil immune system is rapidly evolving to recognize and potentially suppress cancer growth through these pathways. Overall, our study provided necessary data for genomics-based conservation and management in Tasmanian devils.