A primer on the use of machine learning to distil knowledge from data in biological psychiatry.

Applications of machine learning in the biomedical sciences are growing rapidly. This growth has been spurred by diverse cross-institutional and interdisciplinary collaborations, public availability of large datasets, an increase in the accessibility of analytic routines, and the availability of powerful computing resources. With this increased access and exposure to machine learning comes a responsibility for education and a deeper understanding of its bases and bounds, borne equally by data scientists seeking to ply their analytic wares in medical research and by biomedical scientists seeking to harness such methods to glean knowledge from data. This article provides an accessible and critical review of machine learning for a biomedically informed audience, as well as its applications in psychiatry. The review covers definitions and expositions of commonly used machine learning methods, and historical trends of their use in psychiatry. We also provide a set of standards, namely Guidelines for REporting Machine Learning Investigations in Neuropsychiatry (GREMLIN), for designing and reporting studies that use machine learning as a primary data-analysis approach. Lastly, we propose the establishment of the Machine Learning in Psychiatry (MLPsych) Consortium, enumerate its objectives, and identify areas of opportunity for future applications of machine learning in biological psychiatry. This review serves as a cautiously optimistic primer on machine learning for those on the precipice as they prepare to dive into the field, either as methodological practitioners or well-informed consumers.

Genetic and evolutionary divergence of harbour seals (Phoca vitulina) in Iliamna Lake, Alaska.

Freshwater populations of typically marine species present unique opportunities to investigate biodiversity, evolutionary divergence, and the adaptive potential and niche width of species. A few pinniped species have populations that reside solely in freshwater. The harbour seals inhabiting Iliamna Lake, Alaska constitute one such population. Their remoteness, however, has long hindered scientific inquiry. We used DNA from seal scat and tissue samples provided by Indigenous hunters to screen for mitochondrial DNA and microsatellite variation within Iliamna Lake and eight regions across the Pacific Ocean. The Iliamna seals (i) were substantially and significantly discrete from all other populations ( [Formula: see text]F st-mtDNA = 0.544, [Formula: see text]Φ st - mtDNA = 0.541, [Formula: see text]F st-microsatellites = 0.308), (ii) formed a discrete genetic cluster separate from all marine populations (modal ∆k = 2, PC1 = 14.8%), had (iii) less genetic diversity (Hd, π, H exp), and (iv) higher inbreeding (F) than marine populations. These findings are both striking and unexpected revealing that Iliamna seals have likely been on a separate evolutionary trajectory for some time and may represent a unique evolutionary legacy for the species. Attention must now be given to the selective processes driving evolutionary divergence from harbour seals in marine habitats and to ensuring the future of the Iliamna seal.

Distinct evolutionary lineages of Schistocephalus parasites infecting co-occurring sculpin and stickleback fishes in Alaska.

Sculpins (coastrange and slimy) and sticklebacks (ninespine and threespine) are widely distributed fishes cohabiting 2 south-central Alaskan lakes (Aleknagik and Iliamna), and all these species are parasitized by cryptic diphyllobothriidean cestodes in the genus Schistocephalus. The goal of this investigation was to test for host-specific parasitic relationships between sculpins and sticklebacks based upon morphological traits (segment counts) and sequence variation across the NADH1 gene. A total of 446 plerocercoids was examined. Large, significant differences in mean segment counts were found between cestodes in sculpin (mean = 112; standard deviation [s.d.] = 15) and stickleback (mean = 86; s.d. = 9) hosts within and between lakes. Nucleotide sequence divergence between parasites from sculpin and stickleback hosts was 20.5%, and Bayesian phylogenetic analysis recovered 2 well-supported clades of cestodes reflecting intermediate host family (i.e. sculpin, Cottidae vs stickleback, Gasterosteidae). Our findings point to the presence of a distinct lineage of cryptic Schistocephalus in sculpins from Aleknagik and Iliamna lakes that warrants further investigation to determine appropriate evolutionary and taxonomic recognition.

Learning sparse log-ratios for high-throughput sequencing data.

MOTIVATION: The automatic discovery of sparse biomarkers that are associated with an outcome of interest is a central goal of bioinformatics. In the context of high-throughput sequencing (HTS) data, and compositional data (CoDa) more generally, an important class of biomarkers are the log-ratios between the input variables. However, identifying predictive log-ratio biomarkers from HTS data is a combinatorial optimization problem, which is computationally challenging. Existing methods are slow to run and scale poorly with the dimension of the input, which has limited their application to low- and moderate-dimensional metagenomic datasets. RESULTS: Building on recent advances from the field of deep learning, we present CoDaCoRe, a novel learning algorithm that identifies sparse, interpretable and predictive log-ratio biomarkers. Our algorithm exploits a continuous relaxation to approximate the underlying combinatorial optimization problem. This relaxation can then be optimized efficiently using the modern ML toolbox, in particular, gradient descent. As a result, CoDaCoRe runs several orders of magnitude faster than competing methods, all while achieving state-of-the-art performance in terms of predictive accuracy and sparsity. We verify the outperformance of CoDaCoRe across a wide range of microbiome, metabolite and microRNA benchmark datasets, as well as a particularly high-dimensional dataset that is outright computationally intractable for existing sparse log-ratio selection methods. AVAILABILITY AND IMPLEMENTATION: The CoDaCoRe package is available at https://github.com/egr95/R-codacore. Code and instructions for reproducing our results are available at https://github.com/cunningham-lab/codacore. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

GraphDTA: predicting drug-target binding affinity with graph neural networks.

SUMMARY: The development of new drugs is costly, time consuming and often accompanied with safety issues. Drug repurposing can avoid the expensive and lengthy process of drug development by finding new uses for already approved drugs. In order to repurpose drugs effectively, it is useful to know which proteins are targeted by which drugs. Computational models that estimate the interaction strength of new drug-target pairs have the potential to expedite drug repurposing. Several models have been proposed for this task. However, these models represent the drugs as strings, which is not a natural way to represent molecules. We propose a new model called GraphDTA that represents drugs as graphs and uses graph neural networks to predict drug-target affinity. We show that graph neural networks not only predict drug-target affinity better than non-deep learning models, but also outperform competing deep learning methods. Our results confirm that deep learning models are appropriate for drug-target binding affinity prediction, and that representing drugs as graphs can lead to further improvements. AVAILABILITY OF IMPLEMENTATION: The proposed models are implemented in Python. Related data, pre-trained models and source code are publicly available at https://github.com/thinng/GraphDTA. All scripts and data needed to reproduce the post hoc statistical analysis are available from https://doi.org/10.5281/zenodo.3603523. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Implications of Large-Effect Loci for Conservation: A Review and Case Study with Pacific Salmon.

The increasing feasibility of assembling large genomic datasets for non-model species presents both opportunities and challenges for applied conservation and management. A popular theme in recent studies is the search for large-effect loci that explain substantial portions of phenotypic variance for a key trait(s). If such loci can be linked to adaptations, 2 important questions arise: 1) Should information from these loci be used to reconfigure conservation units (CUs), even if this conflicts with overall patterns of genetic differentiation? 2) How should this information be used in viability assessments of populations and larger CUs? In this review, we address these questions in the context of recent studies of Chinook salmon and steelhead (anadromous form of rainbow trout) that show strong associations between adult migration timing and specific alleles in one small genomic region. Based on the polygenic paradigm (most traits are controlled by many genes of small effect) and genetic data available at the time showing that early-migrating populations are most closely related to nearby late-migrating populations, adult migration differences in Pacific salmon and steelhead were considered to reflect diversity within CUs rather than separate CUs. Recent data, however, suggest that specific alleles are required for early migration, and that these alleles are lost in populations where conditions do not support early-migrating phenotypes. Contrasting determinations under the US Endangered Species Act and the State of California's equivalent legislation illustrate the complexities of incorporating genomics data into CU configuration decisions. Regardless how CUs are defined, viability assessments should consider that 1) early-migrating phenotypes experience disproportionate risks across large geographic areas, so it becomes important to identify early-migrating populations that can serve as reliable sources for these valuable genetic resources; and 2) genetic architecture, especially the existence of large-effect loci, can affect evolutionary potential and adaptability.

The maternal gut microbiome during pregnancy and offspring allergy and asthma.

Environmental exposures during pregnancy that alter both the maternal gut microbiome and the infant's risk of allergic disease and asthma include a traditional farm environment and consumption of unpasteurized cow's milk, antibiotic use, dietary fiber, and psychosocial stress. Multiple mechanisms acting in concert may underpin these associations and prime the infant to acquire immune competence and homeostasis following exposure to the extrauterine environment. Cellular and metabolic products of the maternal gut microbiome can promote the expression of microbial pattern recognition receptors, as well as thymic and bone marrow hematopoiesis relevant to regulatory immunity. At birth, transmission of maternally derived bacteria likely leverages this in utero programming to accelerate postnatal transition from a TH2- to TH1- and TH17-dominant immune phenotype and maturation of regulatory immune mechanisms, which in turn reduce the child's risk of allergic disease and asthma. Although our understanding of these phenomena is rapidly evolving, the field is relatively nascent, and we are yet to translate existing knowledge into interventions that substantially reduce disease risk in humans. Here, we review evidence that the maternal gut microbiome impacts the offspring's risk of allergic disease and asthma, discuss challenges and future directions for the field, and propose the hypothesis that maternal carriage of Prevotella copri during pregnancy decreases the offspring's risk of allergic disease via production of succinate, which in turn promotes bone marrow myelopoiesis of dendritic cell precursors in the fetus.

Evolutionary analysis of the Delta and Delta Plus variants of the SARS-CoV-2 viruses.

Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) has been rapidly evolving in the form of new variants. At least eleven known variants have been reported. The objective of this study was to delineate the differences in the mutational profile of Delta and Delta Plus variants. High-quality sequences (n = 1756) of Delta (B.1.617.2) and Delta Plus (AY.1 or B.1.617.2.1) variants were used to determine the prevalence of mutations (≥20 %) in the entire SARS-CoV-2 genome, their co-existence, and change in prevalence over a period of time. Structural analysis was conducted to get insights into the impact of mutations on antibody binding. A Sankey diagram was generated using phylogenetic analysis coupled with sequence-acquisition dates to infer the migration of the Delta Plus variant and its presence in the United States. The Delta Plus variant had a significant number of high-prevalence mutations (≥20 %) than in the Delta variant. Signature mutations in Spike (G142D, A222V, and T95I) existed at a more significant percentage in the Delta Plus variant than the Delta variant. Three mutations in Spike (K417N, V70F, and W258L) were exclusively present in the Delta Plus variant. A new mutation was identified in ORF1a (A1146T), which was only present in the Delta Plus variant with ~58 % prevalence. Furthermore, five key mutations (T95I, A222V, G142D, R158G, and K417N) were significantly more prevalent in the Delta Plus than in the Delta variant. Structural analyses revealed that mutations alter the sidechain conformation to weaken the interactions with antibodies. Delta Plus, which first emerged in India, reached the United States through England and Japan, followed by its spread to more than 20 the United States. Based on the results presented here, it is clear that the Delta and Delta Plus variants have unique mutation profiles, and the Delta Plus variant is not just a simple addition of K417N to the Delta variant. Highly correlated mutations may have emerged to keep the structural integrity of the virus.

Gut Microbiome Diversity and Composition Are Associated with Habitual Dairy Intakes: A Cross-Sectional Study in Men.

BACKGROUND: At a population level, the relation between dairy consumption and gut microbiome composition is poorly understood. OBJECTIVES: We sought to study the cross-sectional associations between individual dairy foods (i.e., milk, yogurt, and cheese), as well as total dairy intake, and the gut microbiome composition in a large, representative sample of men living in south-eastern Australia. METHODS: Data on 474 men (mean ± SD: 64.5 ± 13.5 y old) from the Geelong Osteoporosis Study were used to assess the cross-sectional association between dairy consumption and gut microbiome. Information on dairy intake was self-reported. Men were categorized as consumers and nonconsumers of milk, yogurt, cheese, and high- and low-fat milk. Milk, yogurt, and cheese intakes were summed to calculate the total dairy consumed per day and categorized into either low (<2.5 servings/d) or high (≥2.5 servings/d) total dairy groups. Fecal samples were analyzed using bacterial 16S ribosomal RNA (rRNA) gene sequencing. After assessment of α and ß diversity, differential abundance analysis was performed to identify bacterial taxa associated with each of milk, yogurt, and cheese consumption compared with nonconsumption, low compared with high total dairy, and low- compared with high-fat milk consumption. All analyses were adjusted for potential confounders. RESULTS: α Diversity was not associated with consumption of any of the dairy groups. Differences in ß diversity were observed between milk and yogurt consumption compared with nonconsumption. Taxa belonging to the genera Ruminococcaceae UCG-010 and Bifidobacterium showed negative and weak positive associations with milk consumption, respectively. A taxon from the genus Streptococcus was positively associated with yogurt consumption, whereas a taxon from the genus Eisenbergiella was negatively associated with cheese consumption. No specific taxa were associated with low- compared with high-fat milk nor low compared with high total dairy consumption. CONCLUSIONS: In men, community-level microbiome differences were observed between consumers and nonconsumers of milk and yogurt. Bacterial taxon-level associations were detected with milk, yogurt, and cheese consumption. Total dairy consumption was not associated with any microbiome measures, suggesting that individual dairy foods may have differential roles in shaping the gut microbiome in men.

Deep in the Bowel: Highly Interpretable Neural Encoder-Decoder Networks Predict Gut Metabolites from Gut Microbiome.

BACKGROUND: Technological advances in next-generation sequencing (NGS) and chromatographic assays [e.g., liquid chromatography mass spectrometry (LC-MS)] have made it possible to identify thousands of microbe and metabolite species, and to measure their relative abundance. In this paper, we propose a sparse neural encoder-decoder network to predict metabolite abundances from microbe abundances. RESULTS: Using paired data from a cohort of inflammatory bowel disease (IBD) patients, we show that our neural encoder-decoder model outperforms linear univariate and multivariate methods in terms of accuracy, sparsity, and stability. Importantly, we show that our neural encoder-decoder model is not simply a black box designed to maximize predictive accuracy. Rather, the network's hidden layer (i.e., the latent space, comprised only of sparsely weighted microbe counts) actually captures key microbe-metabolite relationships that are themselves clinically meaningful. Although this hidden layer is learned without any knowledge of the patient's diagnosis, we show that the learned latent features are structured in a way that predicts IBD and treatment status with high accuracy. CONCLUSIONS: By imposing a non-negative weights constraint, the network becomes a directed graph where each downstream node is interpretable as the additive combination of the upstream nodes. Here, the middle layer comprises distinct microbe-metabolite axes that relate key microbial biomarkers with metabolite biomarkers. By pre-processing the microbiome and metabolome data using compositional data analysis methods, we ensure that our proposed multi-omics workflow will generalize to any pair of -omics data. To the best of our knowledge, this work is the first application of neural encoder-decoders for the interpretable integration of multi-omics biological data.

Managing salmon for wildlife: Do fisheries limit salmon consumption by bears in small Alaskan streams?

Ecosystem-based management requires consideration of overlapping resource use between humans and other consumers. Pacific salmon are an important resource for both fisheries and populations of wildlife around the Pacific rim, including coastal brown bears (Ursus arctos); salmon consumption has been positively linked to bear density, body size, and reproductive rate. As a case study within the broader context of human-wildlife competition for food, we used 16-22 yr of empirical data in four different salmon-bearing systems in southwestern Alaska to explore the relationship between sockeye salmon (Oncorhynchus nerka) availability and consumption by bears. We found a negative relationship between the annual biomass of salmon available to bears and the fraction of biomass consumed per fish, and a saturating relationship between salmon availability and the total annual biomass of salmon consumed by bears. Under modeled scenarios, bear consumption of salmon was predicted to increase only with dramatic (on the order of 50-100%) increases in prey availability. Even such large increases in salmon abundance were estimated to produce relatively modest increases in per capita salmon consumption by bears (2.4-4.8 kg·bear-1 ·d-1 , 15-59% of the estimated daily maximum per capita intake), in part because bears did not consume salmon entirely, especially when salmon were most available. Thus, while bears catching salmon in small streams may be limited by salmon harvest in some years, current management of the systems we studied is sufficient for bear populations to reach maximum salmon consumption every 2-4 yr. Consequently, allocating more salmon for brown bear conservation would unlikely result in an ecologically significant response for bears in these systems, though other ecosystem components might benefit. Our results highlight the need for documenting empirical relationships between prey abundance and consumption, particularly in systems with partial consumption, when evaluating the ecological response of managing prey resources for wildlife populations.

Amino acid cues emanating from Pacific salmon eggs and ovarian fluid.

The eggs of salmonid fishes are an important food source for many aquatic predators that detect eggs using olfaction. Moreover, chemicals from eggs and ovarian fluid aid sperm cells in detecting and locating eggs for fertilization, and ovarian fluid is attractive to conspecific males. Thus chemicals from eggs and ovarian fluid may facilitate reproduction but may also attract egg predators. The authors sampled mature females of three Pacific salmon species - Chinook (Oncorhynchus tshawytscha), coho (Oncorhynchus kisutch) and sockeye (Oncorhynchus nerka) - and determined the proportional representation of amino acids, potent fish odorants, from their eggs and ovarian fluid (Chinook and coho salmon only). They then tested juvenile coho salmon, an egg predator, for responses to ovarian fluid and egg odours using the electro-olfactogram (EOG) recording technique. The amino acid compositions of the salmon species were significantly and positively correlated with each other, and the interspecific differences were comparable to those between individuals of the same species. The egg water samples were, on average, dominated by lysine, alanine and glutamine (12.6%, 12.4% and 10.9%, respectively). The ovarian fluid samples were dominated by lysine (20.5%), followed by threonine (9.7%), glycine (9.2%) and arginine (8.8%). EOG recordings demonstrated the ability of juvenile coho salmon to detect the chemical traces of eggs and ovarian fluid. It is concluded that salmon eggs are a potent source of odours for potential predators but likely not highly differentiated among salmon species.

Independent lineages in a common environment: the roles of determinism and contingency in shaping the migration timing of even- versus odd-year pink salmon over broad spatial and temporal scales.

Studies of parallel evolution are seldom able to disentangle the influence of cryptic environmental variation from that of evolutionary history; whereas the unique life history of pink salmon (Oncorhynchus gorbuscha) presents an opportunity to do so. All pink salmon mature at age two and die after breeding. Hence, pink salmon bred in even years are completely reproductively isolated from those bred in odd years, even if the two lineages bred in same location. We used time series (mean = 7 years, maximum = 74 years) of paired even- and odd-year populations from 36 rivers spanning over 2000 km to explore parallelism in migration timing, a trait with a strong genetic basis. Migration timing was highly parallel, being determined almost entirely by local environmental differences among rivers. Interestingly, interannual changes in migration timing different somewhat between lineages. Overall, our findings indicate very strong determinism, with only a minor contribution of contingency.

Understanding sequencing data as compositions: an outlook and review.

Motivation: Although seldom acknowledged explicitly, count data generated by sequencing platforms exist as compositions for which the abundance of each component (e.g. gene or transcript) is only coherently interpretable relative to other components within that sample. This property arises from the assay technology itself, whereby the number of counts recorded for each sample is constrained by an arbitrary total sum (i.e. library size). Consequently, sequencing data, as compositional data, exist in a non-Euclidean space that, without normalization or transformation, renders invalid many conventional analyses, including distance measures, correlation coefficients and multivariate statistical models. Results: The purpose of this review is to summarize the principles of compositional data analysis (CoDA), provide evidence for why sequencing data are compositional, discuss compositionally valid methods available for analyzing sequencing data, and highlight future directions with regard to this field of study. Supplementary information: Supplementary data are available at Bioinformatics online.

Retention of a chromosomal inversion from an anadromous ancestor provides the genetic basis for alternative freshwater ecotypes in rainbow trout.

Migratory behaviour patterns in animals are controlled by a complex genetic architecture. Rainbow trout (Oncorhynchus mykiss) is a salmonid fish that spawns in streams but exhibits three primary life history pathways: stream-resident (fluvial), lake-migrant (adfluvial) and ocean-migrant (anadromous). Previous studies examining fluvial and anadromous O. mykiss have identified several genes associated with life history divergence including the presence of an inversion complex within chromosome 5 (Omy05) that appears to maintain a suite of linked genes controlling migratory behaviour. However, adfluvial trout are migratory without being anadromous, and the genetic basis for this life history has not been investigated from evolutionary perspectives. We sampled wild, native nonanadromous rainbow trout occupying connected stream and lake habitats in a southwest Alaskan watershed to determine whether these fish exhibit genetic divergence between fluvial and adfluvial ecotypes, and whether that divergence parallels that documented in fluvial and anadromous O. mykiss. Data from restriction site-associated DNA (RAD) sequencing revealed an association between frequencies of both the Omy05 inversion complex and other single nucleotide polymorphisms (SNPs) with habitat type (stream or lake), supporting the genetic divergence of fluvial and adfluvial individuals in sympatry. The presence of a genetic basis for migration into lakes, analogous to that documented for anadromy, indicates that the adfluvial ecotype must be recognized separately from the fluvial form of O. mykiss even though neither is anadromous. These results highlight the genetic architecture underlying migration and the importance of chromosomal inversions in promoting and sustaining intraspecific diversity.

Sexual dimorphism modifies habitat-associated divergence: Evidence from beach and creek breeding sockeye salmon.

Studies of parallel or convergent evolution (the repeated, independent evolution of similar traits in similar habitats) rarely explicitly quantify the extent of parallelism (i.e. variation in the direction and/or magnitude of divergence) between the sexes; instead, they often investigate both sexes together or exclude one sex. However, differences in male and female patterns of divergence could contribute to overall variation in the extent of parallelism among ecotype pairs, especially in sexually dimorphic traits. Failing to properly attribute such variation could lead to underestimates of the importance of environmental variation in shaping phenotypes. We investigate the extent of parallelism in the body shape of male and female beach and creek spawning sockeye salmon (Oncorhynchus nerka) from two lake systems in western Alaska that were colonized independently after the last ice age. Although both sexes showed some degree of parallelism, patterns of beach-creek body shape divergence vary between the sexes and between lake systems. Phenotypic change vector analyses revealed highly parallel aspects of divergence between males from different lake systems (males from beaches had deeper bodies than males from creeks) but weaker parallelism in females and high parallelism between the sexes in one lake system but not the other. Body shape also had population-specific components, which were mostly, but not entirely, explained by environmental variation in the form of creek depth. Our results highlight the importance of explicitly considering the extent of parallelism between the sexes and environmental variation among sites within habitat types.

Enhancing the Efficacy of Melanocortin 1 Receptor-Targeted Radiotherapy by Pharmacologically Upregulating the Receptor in Metastatic Melanoma.

Melanocortin 1 receptor (MC1R) is under investigation as a target for drug delivery for metastatic melanoma therapy and imaging. The purpose of this study was to determine the potential of using BRAF inhibitors (BRAFi) and histone deacetylase inhibitors (HDACi) to enhance the delivery of MC1R-targeted radiolabeled peptide ([212Pb]DOTA-MC1L) by pharmacologically upregulating the MC1R expression in metastatic melanoma cells and tumors. MC1R expression was analyzed in de-identified melanoma biopsies by immunohistochemical staining. Upregulation of MC1R expression was determined in BRAFV600E cells (A2058) and BRAF wild-type melanoma cells (MEWO) by quantitative real-time polymerase chain reaction, flow cytometry, and receptor-ligand binding assays. The role of microphthalmia-associated transcription factor (MITF) in the upregulation of MC1R was also examined in A2058 and MEWO cells. The effectiveness of [212Pb]DOTA-MC1L α-particle radiotherapy in combination with BRAFi and/or HDACi was determined in athymic nu/nu mice bearing A2058 and MEWO human melanoma xenografts. High expression of MC1R was observed in situ in clinical melanoma biopsies. BRAFi and HDACi significantly increased the MC1R expression (up to 10-fold in mRNA and 4-fold in protein levels) via MITF-dependent pathways, and this increase led to enhanced ligand binding on the cell surface. Inhibition of MITF expression antagonized the upregulation of MC1R in both BRAFV600E and BRAFWT cells. Combining [212Pb]DOTA-MC1L with BRAFi and/or HDACi improved the tumor response by increasing the delivery of 212Pb α-particle emissions to melanoma tumors via augmented MC1R expression. These data suggest that FDA-approved HDACi and BRAFi could improve the effectiveness of MC1R-targeted therapies by enhancing drug delivery via upregulated MC1R.

A cross-cancer metastasis signature in the microRNA-mRNA axis of paired tissue samples.

In the progression of cancer, cells acquire genetic mutations that cause uncontrolled growth. Over time, the primary tumour may undergo additional mutations that allow for the cancerous cells to spread throughout the body as metastases. Since metastatic development typically results in markedly worse patient outcomes, research into the identity and function of metastasis-associated biomarkers could eventually translate into clinical diagnostics or novel therapeutics. Although the general processes underpinning metastatic progression are understood, no clear cross-cancer biomarker profile has emerged. However, the literature suggests that some microRNAs (miRNAs) may play an important role in the metastatic progression of several cancer types. Using a subset of The Cancer Genome Atlas (TCGA) data, we performed an integrated analysis of mRNA and miRNA expression with paired metastatic and primary tumour samples to interrogate how the miRNA-mRNA regulatory axis influences metastatic progression. From this, we successfully built mRNA- and miRNA-specific classifiers that can discriminate pairs of metastatic and primary samples across 11 cancer types. In addition, we identified a number of miRNAs whose metastasis-associated dysregulation could predict mRNA metastasis-associated dysregulation. Among the most predictive miRNAs, we found several previously implicated in cancer progression, including miR-301b, miR-1296, and miR-423. Taken together, our results suggest that metastatic samples have a common cross-cancer signature when compared with their primary tumour pair, and that these miRNA biomarkers can be used to predict metastatic status as well as mRNA expression.

Spawning and emergence phenology of bull trout Salvelinus confluentus under differing thermal regimes.

Median bull trout Salvelinus confluentus breeding was 2 weeks earlier in a cool stream than in a proximate warmer stream, aligning with expectations for salmonids, followed by emergence timing calculated to be 6 weeks later in the cool stream than the warm stream. This pattern is consistent with both site-specific adaptation and thermal spawning threshold hypotheses for life-history event timing in this threatened species.