A platform-independent AI tumor lineage and site (ATLAS) classifier.

Histopathologic diagnosis and classification of cancer plays a critical role in guiding treatment. Advances in next-generation sequencing have ushered in new complementary molecular frameworks. However, existing approaches do not independently assess both site-of-origin (e.g. prostate) and lineage (e.g. adenocarcinoma) and have minimal validation in metastatic disease, where classification is more difficult. Utilizing gradient-boosted machine learning, we developed ATLAS, a pair of separate AI Tumor Lineage and Site-of-origin models from RNA expression data on 8249 tumor samples. We assessed performance independently in 10,376 total tumor samples, including 1490 metastatic samples, achieving an accuracy of 91.4% for cancer site-of-origin and 97.1% for cancer lineage. High confidence predictions (encompassing the majority of cases) were accurate 98-99% of the time in both localized and remarkably even in metastatic samples. We also identified emergent properties of our lineage scores for tumor types on which the model was never trained (zero-shot learning). Adenocarcinoma/sarcoma lineage scores differentiated epithelioid from biphasic/sarcomatoid mesothelioma. Also, predicted lineage de-differentiation identified neuroendocrine/small cell tumors and was associated with poor outcomes across tumor types. Our platform-independent single-sample approach can be easily translated to existing RNA-seq platforms. ATLAS can complement and guide traditional histopathologic assessment in challenging situations and tumors of unknown primary.

Identification of phenocopies improves prediction of targeted therapy response over DNA mutations alone.

DNA mutations in specific genes can confer preferential benefit from drugs targeting those genes. However, other molecular perturbations can "phenocopy" pathogenic mutations, but would not be identified using standard clinical sequencing, leading to missed opportunities for other patients to benefit from targeted treatments. We hypothesized that RNA phenocopy signatures of key cancer driver gene mutations could improve our ability to predict response to targeted therapies, despite not being directly trained on drug response. To test this, we built gene expression signatures in tissue samples for specific mutations and found that phenocopy signatures broadly increased accuracy of drug response predictions in-vitro compared to DNA mutation alone, and identified additional cancer cell lines that respond well with a positive/negative predictive value on par or better than DNA mutations. We further validated our results across four clinical cohorts. Our results suggest that routine RNA sequencing of tumors to identify phenocopies in addition to standard targeted DNA sequencing would improve our ability to accurately select patients for targeted therapies in the clinic.

Predicting cancer drug TARGETS - TreAtment Response Generalized Elastic-neT Signatures.

We are now in an era of molecular medicine, where specific DNA alterations can be used to identify patients who will respond to specific drugs. However, there are only a handful of clinically used predictive biomarkers in oncology. Herein, we describe an approach utilizing in vitro DNA and RNA sequencing and drug response data to create TreAtment Response Generalized Elastic-neT Signatures (TARGETS). We trained TARGETS drug response models using Elastic-Net regression in the publicly available Genomics of Drug Sensitivity in Cancer (GDSC) database. Models were then validated on additional in-vitro data from the Cancer Cell Line Encyclopedia (CCLE), and on clinical samples from The Cancer Genome Atlas (TCGA) and Stand Up to Cancer/Prostate Cancer Foundation West Coast Prostate Cancer Dream Team (WCDT). First, we demonstrated that all TARGETS models successfully predicted treatment response in the separate in-vitro CCLE treatment response dataset. Next, we evaluated all FDA-approved biomarker-based cancer drug indications in TCGA and demonstrated that TARGETS predictions were concordant with established clinical indications. Finally, we performed independent clinical validation in the WCDT and found that the TARGETS AR signaling inhibitors (ARSI) signature successfully predicted clinical treatment response in metastatic castration-resistant prostate cancer with a statistically significant interaction between the TARGETS score and PSA response (p = 0.0252). TARGETS represents a pan-cancer, platform-independent approach to predict response to oncologic therapies and could be used as a tool to better select patients for existing therapies as well as identify new indications for testing in prospective clinical trials.

The ghosts of propagation past: haplotype information clarifies the relative influence of stocking history and phylogeographic processes on contemporary population structure of walleye (Sander vitreus).

Stocking of fish is an important tool for maintaining fisheries but can also significantly alter population genetic structure and erode the portfolio of within-species diversity that is important for promoting resilience and adaptability. Walleye (Sander vitreus) are a highly valued sportfish in the midwestern United States, a region characterized by postglacial recolonization from multiple lineages and an extensive history of stocking. We leveraged genomic data and recently developed analytical approaches to explore the population structure of walleye from two midwestern states, Minnesota and Wisconsin. We genotyped 954 walleye from 23 populations at ~20,000 loci using genotyping by sequencing and tested for patterns of population structure with single-SNP and microhaplotype data. Populations from Minnesota and Wisconsin were highly differentiated from each other, with additional substructure found in each state. Population structure did not consistently adhere to drainage boundaries, as cases of high intra-drainage and low inter-drainage differentiation were observed. Low genetic structure was observed between populations from the upper Wisconsin and upper Chippewa river watersheds, which are found as few as 50 km apart and were likely homogenized through historical stocking. Nevertheless, we were able to differentiate these populations using microhaplotype-based co-ancestry analysis, providing increased resolution over previous microsatellite studies and our other single SNP-based analyses. Although our results illustrate that walleye population structure has been influenced by past stocking practices, native ancestry still exists in most populations and walleye populations may be able to purge non-native alleles and haplotypes in the absence of stocking. Our study is one of the first to use genomic tools to investigate the influence of stocking on population structure in a nonsalmonid fish and outlines a workflow leveraging recently developed analytical methods to improve resolution of complex population structure that will be highly applicable in many species and systems.

A GT-seq panel for walleye (Sander vitreus) provides important insights for efficient development and implementation of amplicon panels in non-model organisms.

Targeted amplicon sequencing methods, such as genotyping-in-thousands by sequencing (GT-seq), facilitate rapid, accurate, and cost-effective analysis of hundreds of genetic loci in thousands of individuals. Development of GT-seq panels is nontrivial, but studies describing trade-offs associated with different steps of GT-seq panel development are rare. Here, we construct a dual-purpose GT-seq panel for walleye (Sander vitreus), discuss trade-offs associated with different development and genotyping approaches, and provide suggestions for researchers constructing their own GT-seq panels. Our GT-seq panel was developed using an ascertainment set consisting of restriction site-associated DNA data from 954 individuals sampled from 23 populations in Minnesota and Wisconsin, USA. We conducted simulations to test the utility of all loci for parentage analysis and genetic stock identification and designed 600 primer pairs to maximize joint accuracy for these analyses. We then performed three rounds of primer optimization to remove loci that overamplified and our final panel consisted of 436 loci. We also explored different approaches for DNA extraction, multiplexed polymerase chain reaction (PCR) amplification, and cleanup steps during the GT-seq process and discovered the following: (i) inexpensive Chelex extractions performed well for genotyping; (ii) the exonuclease I and shrimp alkaline phosphatase (ExoSAP) procedure included in some current protocols did not improve results substantially and was probably unnecessary; and (iii) it was possible to PCR amplify panels separately and combine them prior to adapter ligation. Well-optimized GT-seq panels are valuable resources for conservation genetics and our findings and suggestions should aid in their construction in myriad taxa.