Genome-scale Datasets Research Articles

Museum skins enable identification of introgression associated with cytonuclear discordance.

Increased sampling of genomes and populations across closely related species has revealed that levels of genetic exchange during and after speciation are higher than previously thought. One obvious manifestation of such exchange is strong cytonuclear discordance, where the divergence in mitochondrial DNA (mtDNA) differs from that for nuclear genes more (or less) than expected from differences between mtDNA and nuclear DNA (nDNA) in population size and mutation rate. Given genome-scale datasets and coalescent modelling, we can now confidently identify cases of strong discordance and test specifically for historical or recent introgression as the cause. Using population sampling, combining exon capture data from historical museum specimens and recently collected tissues we showcase how genomic tools can resolve complex evolutionary histories in the brachyotis group of rock-wallabies (Petrogale). In particular, applying population and phylogenomic approaches we can assess the role of demographic processes in driving complex evolutionary patterns and assess a role of ancient introgression and hybridisation. We find that described species are well supported as monophyletic taxa for nDNA genes, but not for mtDNA, with cytonuclear discordance involving at least four operational taxonomic units (OTUs) across four species which diverged 183-278 kya. ABC modelling of nDNA gene trees supports introgression during or after speciation for some taxon pairs with cytonuclear discordance. Given substantial differences in body size between the species involved, this evidence for gene flow is surprising. Heterogenous patterns of introgression were identified but do not appear to be associated with chromosome differences between species. These and previous results suggest that dynamic past climates across the monsoonal tropics could have promoted reticulation among related species.

AleRax: a tool for gene and species tree co-estimation and reconciliation under a probabilistic model of gene duplication, transfer, and loss.

Genomes are a rich source of information on the pattern and process of evolution across biological scales. How best to make use of that information is an active area of research in phylogenetics. Ideally, phylogenetic methods should not only model substitutions along gene trees, which explain differences between homologous gene sequences, but also the processes that generate the gene trees themselves along a shared species tree. To conduct accurate inferences, one needs to account for uncertainty at both levels, that is, in gene trees estimated from inherently short sequences and in their diverse evolutionary histories along a shared species tree. We present AleRax, a software that can infer reconciled gene trees together with a shared species tree using a simple, yet powerful, probabilistic model of gene duplication, transfer, and loss. A key feature of AleRax is its ability to account for uncertainty in the gene tree and its reconciliation by using an efficient approximation to calculate the joint phylogenetic-reconciliation likelihood and sample reconciled gene trees accordingly. Simulations and analyses of empirical data show that AleRax is one order of magnitude faster than competing gene tree inference tools while attaining the same accuracy. It is consistently more robust than species tree inference methods such as SpeciesRax and ASTRAL-Pro 2 under gene tree uncertainty. Finally, AleRax can process multiple gene families in parallel thereby allowing users to compare competing phylogenetic hypotheses and estimate model parameters, such as duplication, transfer, and loss probabilities for genome-scale datasets with hundreds of taxa. GNU GPL at https://github.com/BenoitMorel/AleRax and data are made available at https://cme.h-its.org/exelixis/material/alerax_data.tar.gz.

Ant backbone phylogeny resolved by modelling compositional heterogeneity among sites in genomic data

Ants are the most ubiquitous and ecologically dominant arthropods on Earth, and understanding their phylogeny is crucial for deciphering their character evolution, species diversification, and biogeography. Although recent genomic data have shown promise in clarifying intrafamilial relationships across the tree of ants, inconsistencies between molecular datasets have also emerged. Here I re-examine the most comprehensive published Sanger-sequencing and genome-scale datasets of ants using model comparison methods that model among-site compositional heterogeneity to understand the sources of conflict in phylogenetic studies. My results under the best-fitting model, selected on the basis of Bayesian cross-validation and posterior predictive model checking, identify contentious nodes in ant phylogeny whose resolution is modelling-dependent. I show that the Bayesian infinite mixture CAT model outperforms empirical finite mixture models (C20, C40 and C60) and that, under the best-fitting CAT-GTR + G4 model, the enigmatic Martialis heureka is sister to all ants except Leptanillinae, rejecting the more popular hypothesis supported under worse-fitting models, that place it as sister to Leptanillinae. These analyses resolve a lasting controversy in ant phylogeny and highlight the significance of model comparison and adequate modelling of among-site compositional heterogeneity in reconstructing the deep phylogeny of insects.

Efficient Bayesian inference under the multispecies coalescent with migration.

Analyses of genome sequence data have revealed pervasive interspecific gene flow and enriched our understanding of the role of gene flow in speciation and adaptation. Inference of gene flow using genomic data requires powerful statistical methods. Yet current likelihood-based methods involve heavy computation and are feasible for small datasets only. Here, we implement the multispecies-coalescent-with-migration model in the Bayesian program bpp, which can be used to test for gene flow and estimate migration rates, as well as species divergence times and population sizes. We develop Markov chain Monte Carlo algorithms for efficient sampling from the posterior, enabling the analysis of genome-scale datasets with thousands of loci. Implementation of both introgression and migration models in the same program allows us to test whether gene flow occurred continuously over time or in pulses. Analyses of genomic data from Anopheles mosquitoes demonstrate rich information in typical genomic datasets about the mode and rate of gene flow.

Detection of trait donors and QTL boundaries for early blight resistance using local ancestry inference in a library of genomic sequences for tomato.

By conducting hierarchical clustering along a sliding window, we generated haplotypes across hundreds of re-sequenced genomes in a few hours. We leveraged our method to define cryptic introgressions underlying disease resistance in tomato (Solanum lycopersicum L.) and to discover resistant germplasm in the tomato seed bank. The genomes of 9 accessions with early blight (Alternaria linariae) disease resistance were newly sequenced and analyzed together with published sequences for 770 tomato and wild species accessions, most of which are available in germplasm collections. Identification of common ancestral haplotypes among resistant germplasm enabled rapid fine mapping of recently discovered quantitative trait loci (QTL) conferring resistance and the identification of possible causal variants. The source of the early blight QTL EB-9 was traced to a vintage tomato named 'Devon Surprise'. Another QTL, EB-5, as well as resistance to bacterial spot disease (Xanthomonas spp.), was traced to Hawaii 7998. A genomic survey of all accessions forecasted EB-9-derived resistance in several heirloom tomatoes, accessions of S. lycopersicum var. cerasiforme, and S. pimpinellifolium PI 37009. Our haplotype-based predictions were validated by screening the accessions against the causal pathogen. There was little evidence of EB-5 prevalence in surveyed contemporary germplasm, presenting an opportunity to bolster tomato disease resistance by adding this rare locus. Our work demonstrates practical insights that can be derived from the efficient processing of large genome-scale datasets, including rapid functional prediction of disease resistance QTL in diverse genetic backgrounds. Finally, our work finds more efficient ways to leverage public genetic resources for crop improvement.

Advances in the reconstruction of the spider tree of life: A roadmap for spider systematics and comparative studies.

In the last decade and a half, advances in genetic sequencing technologies have revolutionized systematics, transforming the field from studying morphological characters or a few genetic markers, to genomic datasets in the phylogenomic era. A plethora of molecular phylogenetic studies on many taxonomic groups have come about, converging on, or refuting prevailing morphology or legacy-marker-based hypotheses about evolutionary affinities. Spider systematics has been no exception to this transformation and the inter-relationships of several groups have now been studied using genomic data. About 51 500 extant spider species have been described, all with a conservative body plan, but innumerable morphological and behavioural peculiarities. Inferring the spider tree of life using morphological data has been a challenging task. Molecular data have corroborated many hypotheses of higher-level relationships, but also resulted in new groups that refute previous hypotheses. In this review, we discuss recent advances in the reconstruction of the spider tree of life and highlight areas where additional effort is needed with potential solutions. We base this review on the most comprehensive spider phylogeny to date, representing 131 of the 132 spider families. To achieve this sampling, we combined six Sanger-based markers with newly generated and publicly available genome-scale datasets. We find that some inferred relationships between major lineages of spiders (such as Austrochiloidea, Palpimanoidea and Synspermiata) are robust across different classes of data. However, several new hypotheses have emerged with different classes of molecular data. We identify and discuss the robust and controversial hypotheses and compile this blueprint to design future studies targeting systematic revisions of these problematic groups. We offer an evolutionary framework to explore comparative questions such as evolution of venoms, silk, webs, morphological traits and reproductive strategies.

Phylogenomic analysis provides diagnostic tools for the identification of Anastrepha fraterculus (Diptera: Tephritidae) species complex.

Insect pests cause tremendous impact to agriculture worldwide. Species identification is crucial for implementing appropriate measures of pest control but can be challenging in closely related species. True fruit flies of the genus Anastrepha Schiner (Diptera: Tephritidae) include some of the most serious agricultural pests in the Americas, with the Anastrepha fraterculus (Wiedemann) complex being one of the most important due to its extreme polyphagy and wide distribution across most of the New World tropics and subtropics. The eight morphotypes described for this complex as well as other closely related species are classified in the fraterculus species group, whose evolutionary relationships are unresolved due to incomplete lineage sorting and introgression. We performed multifaceted phylogenomic approaches using thousands of genes to unravel the evolutionary relationships within the A. fraterculus complex to provide a baseline for molecular diagnosis of these pests. We used a methodology that accommodates variable sources of data (transcriptome, genome, and whole-genome shotgun sequencing) and developed a tool to align and filter orthologs, generating reliable datasets for phylogenetic studies. We inferred 3031 gene trees that displayed high levels of discordance. Nevertheless, the topologies of the inferred coalescent species trees were consistent across methods and datasets, except for one lineage in the A. fraterculus complex. Furthermore, network analysis indicated introgression across lineages in the fraterculus group. We present a robust phylogeny of the group that provides insights into the intricate patterns of evolution of the A. fraterculus complex supporting the hypothesis that this complex is an assemblage of closely related cryptic lineages that have evolved under interspecific gene flow. Despite this complex evolutionary scenario, our subsampling analysis revealed that a set of as few as 80 loci has a similar phylogenetic resolution as the genome-scale dataset, offering a foundation to develop more efficient diagnostic tools in this species group.

Scoutknife: A naïve, whole genome informed phylogenetic robusticity metric

Background: The phylogenetic bootstrap, first proposed by Felsenstein in 1985, is a critically important statistical method in assessing the robusticity of phylogenetic datasets. Core to its concept was the use of pseudo sampling - assessing the data by generating new replicates derived from the initial dataset that was used to generate the phylogeny. In this way, phylogenetic support metrics could overcome the lack of perfect, infinite data. With infinite data, however, it is possible to sample smaller replicates directly from the data to obtain both the phylogeny and its statistical robusticity in the same analysis. Due to the growth of whole genome sequencing, the depth and breadth of our datasets have greatly expanded and are set to only expand further. With genome-scale datasets comprising thousands of genes, we can now obtain a proxy for infinite data. Accordingly, we can potentially abandon the notion of pseudo sampling and instead randomly sample small subsets of genes from the thousands of genes in our analyses. Methods: We introduce Scoutknife, a jackknife-style subsampling implementation that generates 100 datasets by randomly sampling a small number of genes from an initial large-gene dataset to jointly establish both a phylogenetic hypothesis and assess its robusticity. We assess its effectiveness by using 18 previously published datasets and 100 simulation studies. Results: We show that Scoutknife is conservative and informative as to conflicts and incongruence across the whole genome, without the need for subsampling based on traditional model selection criteria. Conclusions: Scoutknife reliably achieves comparable results to selecting the best genes on both real and simulation datasets, while being resistant to the potential biases caused by selecting for model fit. As the amount of genome data grows, it becomes an even more exciting option to assess the robusticity of phylogenetic hypotheses.

Notes from the taxonomic disaster zone: Evolutionary drivers of intractable species boundaries in an Australian lizard clade (Scincidae: Ctenotus).

Genomic-scale datasets, sophisticated analytical techniques, and conceptual advances have disproportionately failed to resolve species boundaries in some groups relative to others. To understand the processes that underlie taxonomic intractability, we dissect the speciation history of an Australian lizard clade that arguably represents a "worst-case" scenario for species delimitation within vertebrates: the Ctenotus inornatus species group, a clade beset with decoupled genetic and phenotypic breaks, uncertain geographic ranges, and parallelism in purportedly diagnostic morphological characters. We sampled hundreds of localities to generate a genomic perspective on population divergence, structure, and admixture. Our results revealed rampant paraphyly of nominate taxa in the group, with lineages that are either morphologically cryptic or polytypic. Isolation-by-distance patterns reflect spatially continuous differentiation among certain pairs of putative species, yet genetic and geographic distances are decoupled in other pairs. Comparisons of mitochondrial and nuclear gene trees, tests of nuclear introgression, and historical demographic modelling identified gene flow between divergent candidate species. Levels of admixture are decoupled from phylogenetic relatedness; gene flow is often higher between sympatric species than between parapatric populations of the same species. Such idiosyncratic patterns of introgression contribute to species boundaries that are fuzzy while also varying in fuzziness. Our results suggest that "taxonomic disaster zones" like the C. inornatus species group result from spatial variation in the porosity of species boundaries and the resulting patterns of genetic and phenotypic variation. This study raises questions about the origin and persistence of hybridizing species and highlights the unique insights provided by taxa that have long eluded straightforward taxonomic categorization.

Congruence between ultraconserved element-based matrices and phylotranscriptomic datasets in the scorpion Tree of Life.

Scorpions are ancient and historically renowned for their potent venom. Traditionally, the systematics of this group of arthropods was supported by morphological characters, until recent phylogenomic analyses (using RNAseq data) revealed most of the higher-level taxa to be non-monophyletic. While these phylogenomic hypotheses are stable for almost all lineages, some nodes have been hard to resolve due to minimal taxonomic sampling (e.g. family Chactidae). In the same line, it has been shown that some nodes in the Arachnid Tree of Life show disagreement between hypotheses generated using transcritptomes and other genomic sources such as the ultraconserved elements (UCEs). Here, we compared the phylogenetic signal of transcriptomes vs. UCEs by retrieving UCEs from new and previously published scorpion transcriptomes and genomes, and reconstructed phylogenies using both datasets independently. We reexamined the monophyly and phylogenetic placement of Chactidae, sampling an additional chactid species using both datasets. Our results showed that both sets of genome-scale datasets recovered highly similar topologies, with Chactidae rendered paraphyletic owing to the placement of Nullibrotheas allenii. As a first step toward redressing the systematics of Chactidae, we establish the family Anuroctonidae (new family) to accommodate the genus Anuroctonus.

Genomic Resources for the First Federally Endangered Lichen: The Florida Perforate Cladonia (Cladonia perforata).

Thirty years after its designation as a federally endangered species, the Florida Perforate Cladonia (FPC) remains imperiled in isolated populations in the Florida scrub in the southeastern USA. For threatened and endangered species, such as FPC, reference genomes provide critical insight into genomic diversity, local adaptations, landscape-level genetics, and phylogenomics. Using high-throughput sequencing, we assemble the first draft nuclear and mitochondrial genomes for the FPC mycobiont-Cladonia perforata. We also assess genetic diversity within and among populations in southeastern Florida using genome-scale data and investigate diversity across the entire nuclear ribosomal cistron, including the standard DNA barcoding marker for fungi. The draft nuclear genome spanned 33.6 Mb, and the complete, circular mitochondrial genome was 59 Kb. We also generated the first chloroplast genome, to our knowledge, for the photobiont genus associated with FPC, an undescribed Asterochloris species. We inferred the presence of multiple, distinct mycobiont parental genotypes (genets) occurring at local scales in southeastern Florida, and strikingly, no genets were shared among even the closest sample sites. All sampled thalli shared identical mitochondrial genomes, while the nuclear ribosomal cistron showed limited variability-highlighting the genetic resolution provided by nuclear genome-scale datasets. The genomic resources generated here provide critical resources for informed conservation efforts for the FPC.

Weaker selection on genes with treatment-specific expression consistent with a limit on plasticity evolution in Arabidopsis thaliana.

Differential gene expression between environments often underlies phenotypic plasticity. However, environment-specific expression patterns are hypothesized to relax selection on genes, and thus limit plasticity evolution. We collated over 27 terabases of RNA-sequencing data on Arabidopsis thaliana from over 300 peer-reviewed studies and 200 treatment conditions to investigate this hypothesis. Consistent with relaxed selection, genes with more treatment-specific expression have higher levels of nucleotide diversity and divergence at nonsynonymous sites but lack stronger signals of positive selection. This result persisted even after controlling for expression level, gene length, GC content, the tissue specificity of expression, and technical variation between studies. Overall, our investigation supports the existence of a hypothesized trade-off between the environment specificity of a gene's expression and the strength of selection on said gene in A. thaliana. Future studies should leverage multiple genome-scale datasets to tease apart the contributions of many variables in limiting plasticity evolution.

A compendium of stable hotspots in the CHO genome.

The use of targeted integration for industrial CHO cell line development currently requires significant upfront effort to identify genomic loci capable of supporting multigram per liter therapeutic protein production from a limited number of transgene copies. To address this barrier to widespread adoption, we characterized transgene expression from thousands of stable hotspots in the CHO genome using the Thousands of Reporters Integrated in Parallel high-throughput screening method. This genome-scale data set was used to define a limited set of epigenetic properties of hotspot regions with sizes on the order of 10 kb. Cell lines with landing pad integrations at eight retargeted hotspot candidates consistently exhibited higher transgene mRNA expression than a commercially viable hotspot in equivalent culture conditions. Initial benchmarking of NISTmAb and trastuzumab productivity from one of these hotspots yielded mAb productivities of approximately 0.7-2 g/L (qP range: 2.9-8.2 pg/cell/day) in small-scale fed-batches. These findings indicate the list of hotspot candidates identified here will be a valuable resource for targeted integration platform development within the CHO community.

Testing Multispecies Coalescent Simulators Using Summary Statistics.

As genomic-scale datasets motivate research on species tree inference, simulators of the multispecies coalescent (MSC) process have become essential for the testing and evaluation of new inference methods. However, the simulators themselves must be tested to ensure that they give valid samples. This work develops methods for checking whether a collection of gene trees is in accord with the MSC model on a given species tree. When applied to well-known simulators, we find that several give flawed samples. The tests presented are capable of validating both topological and metric properties of gene tree samples, and are implemented in a freely available R package MSCsimtester so that developers and users may easily apply them.

Genomes of the cosmopolitan fruit pest Bactrocera dorsalis (Diptera: Tephritidae) reveal its global invasion history and thermal adaptation

IntroductionThe oriental fruit fly Bactrocera dorsalis is one of the most destructive agricultural pests worldwide, with highly debated species delimitation, origin, and global spread routes. ObjectivesOur study intended to (i) resolve the taxonomic uncertainties between B. dorsalis and B. carambolae, (ii) reveal the population structure and global invasion routes of B. dorsalis across Asia, Africa, and Oceania, and (iii) identify genomic regions that are responsible for the thermal adaptation of B. dorsalis. MethodsBased on a high-quality chromosome-level reference genome assembly, we explored the population relationship using a genome-scale single nucleotide polymorphism dataset generated from the resequencing data of 487 B. dorsalis genomes and 25 B. carambolae genomes. Genome-wide association studies and silencing using RNA interference were used to identify and verify the candidate genes associated with extreme thermal stress. ResultsWe showed that B. dorsalis originates from the Southern India region with three independent invasion and spread routes worldwide: (i) from Northern India to Northern Southeast Asia, then to Southern Southeast Asia; (ii) from Northern India to Northern Southeast Asian, then to China and Hawaii; and (iii) from Southern India toward the African mainland, then to Madagascar, which is mainly facilitated by human activities including trade and immigration. Twenty-seven genes were identified by a genome-wide association study to be associated with 11 temperature bioclimatic variables. The Cyp6a9 gene may enhance the thermal adaptation of B. dorsalis and thus boost its invasion, which tended to be upregulated at a hardening temperature of 38 °C. Functional verification using RNA interference silencing against Cyp6a9, led to the specific decrease in Cyp6a9 expression, reducing the survival rate of dsRNA-feeding larvae exposed to extreme thermal stress of 45 °C after heat hardening treatments in B. dorsalis. ConclusionThis study provides insights into the evolutionary history and genetic basis of temperature adaptation in B. dorsalis.

Resolving tricky nodes in the tree of life through amino acid recoding.

Genomic data allowed a detailed resolution of the Tree of Life, but "tricky nodes" such as the root of the animals remain unresolved. Genome-scale datasets are heterogeneous as genes and species are exposed to different pressures, and this can negatively impacts phylogenetic accuracy. We use simulated genomic-scale datasets and show that recoding amino acid data improves accuracy when the model does not account for the compositional heterogeneity of the amino acid alignment. We apply our findings to three datasets addressing the root of the animal tree, where the debate centers on whether sponges (Porifera) or comb jellies (Ctenophora) represent the sister of all other animals. We show that results from empirical data follow predictions from simulations and suggest that, at the least in phylogenies inferred from amino acid sequences, a placement of the ctenophores as sister to all the other animals is best explained as a tree reconstruction artifact.

A Practical Guide to Design and Assess a Phylogenomic Study.

Over the last decade, molecular systematics has undergone a change of paradigm as high-throughput sequencing now makes it possible to reconstruct evolutionary relationships using genome-scale datasets. The advent of “big data” molecular phylogenetics provided a battery of new tools for biologists but simultaneously brought new methodological challenges. The increase in analytical complexity comes at the price of highly specific training in computational biology and molecular phylogenetics, resulting very often in a polarized accumulation of knowledge (technical on one side and biological on the other). Interpreting the robustness of genome-scale phylogenetic studies is not straightforward, particularly as new methodological developments have consistently shown that the general belief of “more genes, more robustness” often does not apply, and because there is a range of systematic errors that plague phylogenomic investigations. This is particularly problematic because phylogenomic studies are highly heterogeneous in their methodology, and best practices are often not clearly defined. The main aim of this article is to present what I consider as the ten most important points to take into consideration when planning a well-thought-out phylogenomic study and while evaluating the quality of published papers. The goal is to provide a practical step-by-step guide that can be easily followed by nonexperts and phylogenomic novices in order to assess the technical robustness of phylogenomic studies or improve the experimental design of a project.

Beetle bioluminescence outshines extant aerial predators.

We understand very little about the timing and origins of bioluminescence, particularly as a predator avoidance strategy. Understanding the timing of its origins, however, can help elucidate the evolution of this ecologically important signal. Using fireflies, a prevalent bioluminescent group where bioluminescence primarily functions as aposematic and sexual signals, we explore the origins of this signal in the context of their potential predators. Divergence time estimations were performed using genomic-scale datasets providing a robust estimate for the origin of firefly bioluminescence as both a terrestrial and as an aerial signal. Our results recover the origin of terrestrial beetle bioluminescence at 141.17 (122.63-161.17) Ma and firefly aerial bioluminescence at 133.18 (117.86-152.47) Ma using a large dataset focused on Lampyridae; and terrestrial bioluminescence at 148.03 (130.12-166.80) Ma, with the age of aerial bioluminescence at 104.97 (99.00-120.90) Ma using a complementary Elateroidea dataset. These ages pre-date the origins of all known extant aerial predators (i.e. bats and birds) and support much older terrestrial predators (assassin bugs, frogs, ground beetles, lizards, snakes, hunting spiders and harvestmen) as the drivers of terrestrial bioluminescence in beetles. These ages also support the hypothesis that sexual signalling was probably the original function of this signal in aerial fireflies.

Systematic inference of indirect transcriptional regulation by protein kinases and phosphatases.

Gene expression is controlled by pathways of regulatory factors often involving the activity of protein kinases on transcription factor proteins. Despite this well established mechanism, the number of well described pathways that include the regulatory role of protein kinases on transcription factors is surprisingly scarce in eukaryotes.To address this, PhosTF was developed to infer functional regulatory interactions and pathways in both simulated and real biological networks, based on linear cyclic causal models with latent variables. GeneNetWeaverPhos, an extension of GeneNetWeaver, was developed to allow the simulation of perturbations in known networks that included the activity of protein kinases and phosphatases on gene regulation. Over 2000 genome-wide gene expression profiles, where the loss or gain of regulatory genes could be observed to perturb gene regulation, were then used to infer the existence of regulatory interactions, and their mode of regulation in the budding yeast Saccharomyces cerevisiae.Despite the additional complexity, our inference performed comparably to the best methods that inferred transcription factor regulation assessed in the DREAM4 challenge on similar simulated networks. Inference on integrated genome-scale data sets for yeast identified ∼ 8800 protein kinase/phosphatase-transcription factor interactions and ∼ 6500 interactions among protein kinases and/or phosphatases. Both types of regulatory predictions captured statistically significant numbers of known interactions of their type. Surprisingly, kinases and phosphatases regulated transcription factors by a negative mode or regulation (deactivation) in over 70% of the predictions.

Abstract 1222: Efficient domain adaptation for cancer mutation prediction

Abstract Prediction models have been widely used for many purposes in cancer research, including calling mutation status, identifying cancer subtype, and performing prognostic analysis. A fundamental assumption in supervised machine learning is that the data to be classified is derived from the same distribution as the data used to train the classifier. However, challenges in data acquisition often mean few or no labeled examples are available for the distribution of interest. For instance, sample sizes may be insufficient to train on rare cancer types, and technological limitations can hinder label generation, for instance a lack of simultaneous profiling of gene expression and mutation information in single-cell data. For such situations where labeled target data is limited, the field of domain adaptation and transfer learning has established principled ways to develop predictors for the data of interest (target data) using labeled data from a similar but distinct distribution (source data). One recent method, weighted elastic net domain adaptation (wenda), leverages the complex interactions between features (such as genes) to optimize a model’s predictive power on both source and target datasets. It learns the dependency structure between features and prioritizes those that are similar across distributions. This has previously been shown to significantly improve accuracy on predictions from a mismatched distribution, overcoming the limitations of traditional supervised models. Unfortunately, wenda requires training a Gaussian process model for each feature separately, which is computationally expensive and resists parallelization, making it infeasible for researchers to use at genome-scale. We have developed and implemented a modified form of the underlying algorithm, called wenda_gpu, which allows for fast, efficient model training for genome-scale datasets on a single GPU-enabled computer. Our implementation exploits both quasi-Newtonian parameter optimization and the computational power of GPUs for significant speedups without sacrificing accuracy. Our implementation is able to tackle training tasks on data at the scale of The Cancer Genome Atlas (TCGA), which was infeasible without our technical advances. We demonstrate the use of wenda_gpu on a range of TCGA-scale prediction tasks, making it possible to build accurate, predictive models that generalize to target datasets where supervised models could not be trained due to the lack of labeled data. We also trained models from gene expression data for cross-cancer type mutation prediction, which outperformed a regular elastic net. We anticipate that wenda_gpu will enable researchers to build accurate predictive models in cases where supervised models were previously not possible due to lack of labeled data, including rare cancers. Citation Format: Ariel A. Hippen, Jake Crawford, Jacob R. Gardner, Casey S. Greene. Efficient domain adaptation for cancer mutation prediction [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1222.