Search

Patient-derived xenografts (PDX) are tumor-in-mouse models for cancer. PDX collections, such as the NCI PDXNet, are powerful resources for preclinical therapeutic testing. However, variations in experimental and analysis procedures have limited interpretability. To determine the robustness of PDX studies, the PDXNet tested temozolomide drug response for three prevalidated PDX models (sensitive, resistant, and intermediate) across four blinded PDX Development and Trial Centers using independently selected standard operating procedures. Each PDTC was able to correctly identify the sensitive, resistant, and intermediate models, and statistical evaluations were concordant across all groups. We also developed and benchmarked optimized PDX informatics pipelines, and these yielded robust assessments across xenograft biological replicates. These studies show that PDX drug responses and sequence results are reproducible across diverse experimental protocols. In addition, we share the range of experimental procedures that maintained robustness, as well as standardized cloud-based workflows for PDX exome-sequencing and RNA-sequencing analyses and for evaluating growth.Significance:The PDXNet Consortium shows that PDX drug responses and sequencing results are reproducible across diverse experimental protocols, establishing the potential for multisite preclinical studies to translate into clinical trials.

To cure a patient's cancer is to eradicate invasive cells from the ecosystem of the body. However, the ecologic complexity of this challenge is not well understood. Here we show how results from eradications of invasive mammalian species from islands—one of the few contexts in which invasive species have been regularly cleared—inform new research directions for treating cancer. We first summarize the epidemiologic characteristics of island invader eradications and cancer treatments by analyzing recent datasets from the Database of Invasive Island Species Eradications and The Cancer Genome Atlas, detailing the superior successes of island eradication projects. Next, we compare how genetic and environmental factors impact success in each system. These comparisons illuminate a number of promising cancer research and treatment directions, such as heterogeneity engineering as motivated by gene drives and adaptive therapy; multiscale analyses of how population heterogeneity potentiates treatment resistance; and application of ecological data mining techniques to high-throughput cancer data. We anticipate that interdisciplinary comparisons between tumor progression and invasive species would inspire development of novel paradigms to cure cancer.

The single-molecule multiplex chromatin interaction data are generated by emerging 3D genome mapping technologies such as GAM, SPRITE, and ChIA-Drop. These datasets provide insights into high-dimensional chromatin organization, yet introduce new computational challenges. Thus, we developed MIA-Sig, an algorithmic solution based on signal processing and information theory. We demonstrate its ability to de-noise the multiplex data, assess the statistical significance of chromatin complexes, and identify topological domains and frequent inter-domain contacts. On chromatin immunoprecipitation (ChIP)-enriched data, MIA-Sig can clearly distinguish the protein-associated interactions from the non-specific topological domains. Together, MIA-Sig represents a novel algorithmic framework for multiplex chromatin interaction analysis.

Cancer progression is an evolutionary process. During this process, evolving cancer cell populations encounter restrictive ecological niches within the body, such as the primary tumor, circulatory system, and diverse metastatic sites. Efforts to prevent or delay cancer evolution—and progression—require a deep understanding of the underlying molecular evolutionary processes. Herein we discuss a suite of concepts and tools from evolutionary and ecological theory that can inform cancer biology in new and meaningful ways. We also highlight current challenges to applying these concepts, and propose ways in which incorporating these concepts could identify new therapeutic modes and vulnerabilities in cancer.

Introduction Presently, programmed death ligand 1 is the most commonly used biomarker to predict response to immune checkpoint inhibitors (ICIs) in NSCLC. Owing to its several limitations, there is continuous search for more precise and reliable markers. Frameshift mutations by insertion or deletion (fsindels) are suggested to induce more immunogenic tumor-specific neoantigens, conferring better response to ICIs. Positive correlation of fsindels with ICI response has been studied in melanoma and renal cell carcinoma. We investigated the implication of fsindels in the clinical outcomes and immune landscape of patients with NSCLC treated with ICIs. Methods We utilized The Cancer Genome Atlas data set to analyze tumor mutational burden, neoantigen burden, and immune landscape in relation to fsindel status. In addition, utilizing the clinical data from 122 patients treated with ICIs, we evaluated the influence of fsindels on disease response rates and survival outcomes. Results A positive correlation between fsindel burden and tumor mutational burden and activated CD4/CD8 T-cell infiltration was shown. Presence of fsindels was also associated with significant prolongation of progression-free survival in patients treated with ICIs (median 6.2 versus 2.7 months [p = 0.01]). In addition, significant differences in the overall response rates (26% versus 12% [p = 0.04]) and disease control rates (68% versus 48% [p = 0.02]) were observed in patients with fsindels. Conclusion Our findings suggest that fsindels may have a predictive role for ICI response in NSCLC.

Colonization of visceral organs with melanoma in humanized NSG-SGM3 mice is dependent upon human CD33+CD11b+CD117+ progenitor cells. A gene signature of KIT/CD117–expressing CD33+ subset correlates with decreased overall survival in TCGA melanoma samples and represents a novel candidate biomarker., Metastasis of melanoma significantly worsens prognosis; thus, therapeutic interventions that prevent metastasis could improve patient outcomes. Here, we show using humanized mice that colonization of distant visceral organs with melanoma is dependent upon a human CD33+CD11b+CD117+ progenitor cell subset comprising, Graphical Abstract

Deep learning models are extensively used for analyzing hematoxylin and eosin stained whole slide images. They enjoy high prediction accuracies, but may suffer large performance drops when applied to out-of-sample data. Here we systematically investigate how resolution differences between train and test sets may affect lung cancer subtype predictions from whole slide images using a transfer learning model based on the Inception V3 network. We observe models trained on blurred images perform well when applied to test sets with similar blurrings, but suffer poor predictions when applied to images with large blurring differences. In particular, we observed low area under curve values when models trained on blurred images were applied to non-blurred images.

Ribosome-associated quality control pathways respond to defects in translational elongation to recycle arrested ribosomes and degrade aberrant polypeptides and mRNAs. Loss of a tRNA gene leads to ribosomal pausing that is resolved by the translational GTPase GTPBP2, and in its absence causes neuron death. Here, we show that loss of the homologous protein GTPBP1 during tRNA deficiency in the mouse brain also leads to codon-specific ribosome pausing and neurodegeneration, suggesting that these non-redundant GTPases function in the same pathway to mitigate ribosome pausing. As observed in Gtpbp2-/- mice (Ishimura et al., 2016), GCN2-mediated activation of the integrated stress response (ISR) was apparent in the Gtpbp1-/- brain. We observed decreased mTORC1 signaling which increased neuronal death, whereas ISR activation was neuroprotective. Our data demonstrate that GTPBP1 functions as an important quality control mechanism during translation elongation and suggest that translational signaling pathways intricately interact to regulate neuronal homeostasis during defective elongation.

Background Tumors with high tumor mutational burden (TMB) or defects in mismatch repair (dMMR) respond well to immune checkpoint inhibitors (ICIs).1 2 TMB and DNA repair gene mutations including dMMR are closely related to the increase of neoantigens, which are recognized by immune cells to trigger an immune response.1 3 Although not a standard of care in thyroid cancer treatment, there are ongoing clinical trials for ICI use in differentiated thyroid carcinoma. However, not much has been explored concerning the neoantigen landscape and its association with immune traits in papillary thyroid cancer (PTC). We aim to analyze the immune landscape of PTC in association with neoantigen burden, TMB, and DNA repair gene mutations. Methods We used the PTC cohort data from The Cancer Genome Atlas (TCGA). The mutation counts and data for neoantigen prediction were acquired from TCGA mutation calling. CloudNeo pipeline was used for neoantigen prediction. TMB was calculated as the sum of missense and indel mutation counts per megabase pairs covered by whole-exome sequencing. Tumor-infiltrating immune cells were estimated using CIBERSORT. Results Out of the 496 PTC patients from cBioPortal, a subset of 400 patients with available mutation counts and predicted neoantigen burden was included in the study. Immune cell infiltration estimated by CIBERSORT showed macrophage M2 as the most abundant, followed by macrophage M0 and other T cells (figure 1). The TMB ranged from 0.03 to 2.05 with a median value of 0.2. Neoantigen burden ranged from 0 to 18 with a median value of 1, which is relatively low compared to the median value of 18 in non-small cell lung cancer (NSCLC)1 (figure 2). One or more DNA repair gene mutations were discovered in 32 patients (8%). The mutation status of repair genes was not related to TMB or neoantigen burden. TMB or neoantigen burden was not related to immune traits such as infiltration of CD8+ T cells or regulatory T cells, cytolytic activity score, and PD-L1 expression. Conclusions This is the first study to report the immune landscape of PTC in the context of neoantigen. The lack of association between TMB or neoantigen burden with immune traits may be due to the relatively low number of neoantigens in PTC compared to other immunogenic cancers such as NSCLC. Our results suggest that mutations in DNA repair genes or TMB are likely to have limited value in predicting response to ICI treatment in PTC. References Chae YK, et al., Mutations in DNA repair genes are associated with increased neoantigen burden and a distinct immunophenotype in lung squamous cell carcinoma. Sci Rep 2019; 9:3235. Rizvi NA, et al., Cancer immunology. mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung cancer. Science 2015; 348:124–128. Schumacher TN, Schreiber RD, Neoantigens in cancer immunotherapy. Science 2015; 348:69–74.

Deep learning has recently become a key methodology for the study and interpretation of cancer histology images. The ability of convolutional neural networks (CNNs) to automatically learn features from raw data without the need for pathologist expert knowledge, as well as the availability of annotated histopathology datasets, have contributed to a growing interest in deep learning applications to histopathology. In clinical practice for cancer, histopathological images have been commonly used for diagnosis, prognosis, and treatment. Recently, molecular subtype classification has gained significant attention for predicting standard chemotherapy's outcomes and creating personalized targeted cancer therapy. Genomic profiles, especially gene expression data, are mostly used for molecular subtyping. In this study, we developed a novel, PanCancer CNN model based on Google Inception V3 transfer learning to classify molecular subtypes using histopathological images. We used 22,484 Haemotoxylin and Eosin (HE macro-average=0.90). In cancer studies, combining histopathological images with genomic data has rarely been explored. We investigated the relationship between features extracted from HE macro-average=0.97). These results show that integrating H&E images and gene expression profiles can enhance accuracy of molecular subtype classification.

The evolutionary processes that drive universal therapeutic resistance in adult patients with diffuse glioma remain unclear1,2. Here we analysed temporally separated DNA-sequencing data and matched clinical annotation from 222 adult patients with glioma. By analysing mutations and copy numbers across the three major subtypes of diffuse glioma, we found that driver genes detected at the initial stage of disease were retained at recurrence, whereas there was little evidence of recurrence-specific gene alterations. Treatment with alkylating agents resulted in a hypermutator phenotype at different rates across the glioma subtypes, and hypermutation was not associated with differences in overall survival. Acquired aneuploidy was frequently detected in recurrent gliomas and was characterized by IDH mutation but without co-deletion of chromosome arms 1p/19q, and further converged with acquired alterations in the cell cycle and poor outcomes. The clonal architecture of each tumour remained similar over time, but the presence of subclonal selection was associated with decreased survival. Finally, there were no differences in the levels of immunoediting between initial and recurrent gliomas. Collectively, our results suggest that the strongest selective pressures occur during early glioma development and that current therapies shape this evolution in a largely stochastic manner.

Background Triple negative breast cancer (TNBC) is aggressive with limited treatment options upon recurrence. Molecular discordance between primary and metastatic TNBC has been observed, but the degree of biological heterogeneity has not been fully explored. Furthermore, genomic evolution through treatment is poorly understood. In this study, we aim to characterize the genomic changes between paired primary and metastatic TNBCs through transcriptomic and genomic profiling, and to identify genomic alterations which may contribute to chemotherapy resistance. Methods Genomic alterations and mRNA expression of 10 paired primary and metastatic TNBCs were determined through targeted sequencing, microarray analysis, and RNA sequencing. Commonly mutated genes, as well as differentially expressed and co-expressed genes were identified. We further explored the clinical relevance of differentially expressed genes between primary and metastatic tumors to patient survival using large public datasets. Results Through gene expression profiling, we observed a shift in TNBC subtype classifications between primary and metastatic TNBCs. A panel of eight cancer driver genes (CCNE1, TPX2, ELF3, FANCL, JAK2, GSK3B, CEP76, and SYK) were differentially expressed in recurrent TNBCs, and were also overexpressed in TCGA and METABRIC. CCNE1 and TPX2 were co-overexpressed in TNBCs. DNA mutation profiling showed that multiple mutations occurred in genes comprising a number of potentially targetable pathways including PI3K/AKT/mTOR, RAS/MAPK, cell cycle, and growth factor receptor signaling, reaffirming the wide heterogeneity of mechanisms driving TNBC. CCNE1 amplification was associated with poor overall survival in patients with metastatic TNBC. Conclusions CCNE1 amplification may confer resistance to chemotherapy and is associated with poor overall survival in TNBC. Electronic supplementary material The online version of this article (10.1186/s12885-019-5290-4) contains supplementary material, which is available to authorized users.

Histopathological images are a rich but incompletely explored data type for studying cancer. Manual inspection is time consuming, making it challenging to use for image data mining. Here we show that convolutional neural networks (CNNs) can be systematically applied across cancer types, enabling comparisons to reveal shared spatial behaviors. We develop CNN architectures to analyze 27,815 hematoxylin and eosin scanned images from The Cancer Genome Atlas for tumor/normal, cancer subtype, and mutation classification. Our CNNs are able to classify TCGA pathologist-annotated tumor/normal status of whole slide images (WSIs) in 19 cancer types with consistently high AUCs (0.995 ± 0.008), as well as subtypes with lower but significant accuracy (AUC 0.87 ± 0.1). Remarkably, tumor/normal CNNs trained on one tissue are effective in others (AUC 0.88 ± 0.11), with classifier relationships also recapitulating known adenocarcinoma, carcinoma, and developmental biology. Moreover, classifier comparisons reveal intra-slide spatial similarities, with an average tile-level correlation of 0.45 ± 0.16 between classifier pairs. Breast cancers, bladder cancers, and uterine cancers have spatial patterns that are particularly easy to detect, suggesting these cancers can be canonical types for image analysis. Patterns for TP53 mutations can also be detected, with WSI self- and cross-tissue AUCs ranging from 0.65-0.80. Finally, we comparatively evaluate CNNs on 170 breast and colon cancer images with pathologist-annotated nuclei, finding that both cellular and intercellular regions contribute to CNN accuracy. These results demonstrate the power of CNNs not only for histopathological classification, but also for cross-comparisons to reveal conserved spatial behaviors across tumors., Histopathological images are a rich but incompletely explored data type for studying cancer. Here the authors show that convolutional neural networks can be systematically applied across cancer types, enabling comparisons to reveal shared spatial behaviors.

Patient-Derived Xenografts (PDXs) are tumor-in-mouse models for cancer. PDX collections, such as those supported by the NCI PDXNet program, are powerful resources for preclinical therapeutic testing. However, variations in experimental design and analysis procedures have limited interpretability. To determine the robustness of PDX studies, the PDXNet tested temozolomide drug response for three pre-validated PDX models (sensitive, resistant, and intermediate) across four blinded PDX Development and Trial Centers (PDTCs) using independently selected SOPs. Each PDTC was able to correctly identify the sensitive, resistant, and intermediate models, and statistical evaluations were concordant across all groups. We also developed and benchmarked optimized PDX informatics pipelines, and these yielded robust assessments across xenograft biological replicates. These studies show that PDX drug responses and sequence results are reproducible across diverse experimental protocols. Here we share the range of experimental procedures that maintained robustness, as well as standardized cloud-based workflows for PDX exome-seq and RNA-Seq analysis and for evaluating growth.

MotivationModern genomic research relies heavily on next-generation sequencing experiments such as ChIP-seq and ChIA-PET that generate coverage files for transcription factor binding, as well as DHS and ATAC-seq that yield coverage files for chromatin accessibility. Such files are in a bedGraph text format or a bigWig binary format. Obtaining summary statistics in a given region is a fundamental task in analyzing protein binding intensity or chromatin accessibility. However, the existing Python package for operating on coverage files is not optimized for speed.ResultsWe developed pyBedGraph, a Python package to quickly obtain summary statistics for a given interval in a bedGraph file. When tested on 8 ChIP-seq and ATAC-seq datasets, pyBedGraph is on average 245 times faster than the existing program. Notably, pyBedGraph can look up the exact mean signal of 1 million regions in ~0.26 second on a conventional laptop. An approximate mean for 10,000 regions can be computed in ~0.0012 second with an error rate of less than 5 percent.AvailabilitypyBedGraph is publicly available at https://github.com/TheJacksonLaboratory/pyBedGraph under the MIT license.

Background Patient-derived xenograft (PDX) models are in vivo models of human cancer that have been used for translational cancer research and therapy selection for individual patients. The Jackson Laboratory (JAX) PDX resource comprises 455 models originating from 34 different primary sites (as of 05/08/2019). The models undergo rigorous quality control and are genomically characterized to identify somatic mutations, copy number alterations, and transcriptional profiles. Bioinformatics workflows for analyzing genomic data obtained from human tumors engrafted in a mouse host (i.e., Patient-Derived Xenografts; PDXs) must address challenges such as discriminating between mouse and human sequence reads and accurately identifying somatic mutations and copy number alterations when paired non-tumor DNA from the patient is not available for comparison. Results We report here data analysis workflows and guidelines that address these challenges and achieve reliable identification of somatic mutations, copy number alterations, and transcriptomic profiles of tumors from PDX models that lack genomic data from paired non-tumor tissue for comparison. Our workflows incorporate commonly used software and public databases but are tailored to address the specific challenges of PDX genomics data analysis through parameter tuning and customized data filters and result in improved accuracy for the detection of somatic alterations in PDX models. We also report a gene expression-based classifier that can identify EBV-transformed tumors. We validated our analytical approaches using data simulations and demonstrated the overall concordance of the genomic properties of xenograft tumors with data from primary human tumors in The Cancer Genome Atlas (TCGA). Conclusions The analysis workflows that we have developed to accurately predict somatic profiles of tumors from PDX models that lack normal tissue for comparison enable the identification of the key oncogenic genomic and expression signatures to support model selection and/or biomarker development in therapeutic studies. A reference implementation of our analysis recommendations is available at https://github.com/TheJacksonLaboratory/PDX-Analysis-Workflows. Electronic supplementary material The online version of this article (10.1186/s12920-019-0551-2) contains supplementary material, which is available to authorized users.

Despite advances in therapy, glioblastoma remains an incurable disease with a dismal prognosis. Recent studies have implicated cancer stem cells within glioblastoma (glioma stem cells, GSCs) as mediators of therapeutic resistance and tumor progression. In this study, we investigated the role of the transforming growth factor-β (TGF-β) superfamily, which has been found to play an integral role in the maintenance of stem cell homeostasis within multiple stem cell systems, as a mediator of stem-like cells in glioblastoma. We find that BMP and TGF-β signaling define divergent molecular and functional identities in glioblastoma, and mark relatively quiescent and proliferative GSCs, respectively. Treatment of GSCs with BMP inhibits cell proliferation, but does not abrogate their stem-ness, as measured by self-renewal and tumorigencity. Further, BMP pathway activation confers relative resistance to radiation and temozolomide chemotherapy. Our findings define a quiescent cancer stem cell population in glioblastoma that may be a cellular reservoir for tumor recurrence following cytotoxic therapy.

The single-molecule multiplex chromatin interaction data generated by emerging non-ligation-based 3D genome mapping technologies provide novel insights into high dimensional chromatin organization, yet introduce new computational challenges. We developed MIA-Sig (https://github.com/TheJacksonLaboratory/mia-sig.git), an algorithmic framework to de-noise the data, assess the statistical significance of chromatin complexes, and identify topological domains and inter-domain contacts. On chromatin immunoprecipitation (ChIP)-enriched data, MIA-Sig can clearly distinguish the protein-associated interactions from the non-specific topological domains.

Bioinformatics has become an indispensable part of life science over the past 2 decades. However, bioinformatics education is not well integrated at the undergraduate level, especially in liberal arts colleges and regional universities in the United States. One significant obstacle pointed out by the Network for Integrating Bioinformatics into Life Sciences Education is the lack of faculty in the bioinformatics area. Most current life science professors did not acquire bioinformatics analysis skills during their own training. Consequently, a great number of undergraduate and graduate students do not get the chance to learn bioinformatics or computational biology skills within a structured curriculum during their education. To address this gap, we developed a module-based, week-long short course to train small college and regional university professors with essential bioinformatics skills. The bioinformatics modules were built to be adapted by the professor-trainees afterward and used in their own classes. All the course materials can be accessed at https://github.com/TheJacksonLaboratory/JAXBD2K-ShortCourse.

Deficiencies in DNA repair pathways, including mismatch repair (MMR), have been linked to higher tumor mutation burden and improved response to immune checkpoint inhibitors. However, the significance of MMR mutations in lung cancer has not been well characterized, and the relevance of other processes, including homologous recombination (HR) and polymerase epsilon (POLE) activity, remains unclear. Here, we analyzed a dataset of lung squamous cell carcinoma samples from The Cancer Genome Atlas. Variants in DNA repair genes were associated with increased tumor mutation and neoantigen burden, which in turn were linked with greater tumor infiltration by activated T cells. The subset of tumors with DNA repair gene variants but without T cell infiltration exhibited upregulation of TGF-β and Wnt pathway genes, and a combined score incorporating these genes and DNA repair status accurately predicted immune cell infiltration. Finally, high neoantigen burden was positively associated with genes related to cytolytic activity and immune checkpoints. These findings provide evidence that DNA repair pathway defects and immunomodulatory genes together lead to specific immunophenotypes in lung squamous cell carcinoma and could potentially serve as biomarkers for immunotherapy.

Purpose: Genomic and transcriptomic alterations during metastasis are considered to affect clinical outcome of colorectal cancers, but detailed clinical implications of metastatic alterations are not fully uncovered. We aimed to investigate the effect of metastatic evolution on in vivo treatment outcome, and identify genomic and transcriptomic alterations associated with drug responsiveness. Experimental Design: We developed and analyzed patient-derived xenograft (PDX) models from 35 patients with colorectal cancer including 5 patients with multiple organ metastases (MOMs). We performed whole-exome, DNA methylation, and RNA sequencing for patient and PDX tumors. With samples from patients with MOMs, we conducted phylogenetic and subclonal analysis and in vivo drug efficacy test on the corresponding PDX models. Results: Phylogenetic analysis using mutation, expression, and DNA methylation data in patients with MOMs showed that mutational alterations were closely connected with transcriptomic and epigenomic changes during the tumor evolution. Subclonal analysis revealed that initial primary tumors with larger number of subclones exhibited more dynamic changes in subclonal architecture according to metastasis, and loco-regional and distant metastases occurred in a parallel or independent fashion. The PDX models from MOMs demonstrated therapeutic heterogeneity for targeted treatment, due to subclonal acquisition of additional mutations or transcriptomic activation of bypass signaling pathway during tumor evolution. Conclusions: This study demonstrated in vivo therapeutic heterogeneity of colorectal cancers using PDX models, and suggests that acquired subclonal alterations in mutations or gene expression profiles during tumor metastatic processes can be associated with the development of drug resistance and therapeutic heterogeneity of colorectal cancers.

Bioinformatics workflows for analyzing genomic data obtained from xenografted tumor (e.g., human tumors engrafted in a mouse host) must address several challenges, including separating mouse and human sequence reads and accurate identification of somatic mutations and copy number aberrations when paired normal DNA from the patient is not available. We report here data analysis workflows that address these challenges and result in reliable identification of somatic mutations, copy number alterations, and transcriptomic profiles of tumors from patient derived xenograft models. We validated our analytical approaches using simulated data and by assessing concordance of the genomic properties of xenograft tumors with data from primary human tumors in The Cancer Genome Atlas (TCGA). The commands and parameters for the workflows are available at https://github.com/TheJacksonLaboratory/PDX-Analysis-Workflows.

Mutant allele frequency distributions in cancer samples have been used to estimate intratumoral heterogeneity and its implications for patient survival. However, mutation calls are sensitive to the calling algorithm. It remains unknown whether the relationship of heterogeneity and clinical outcome is robust to these variations. To resolve this question, we studied the robustness of allele frequency distributions to the mutation callers MuTect, SomaticSniper, and VarScan in 4722 cancer samples from The Cancer Genome Atlas. We observed discrepancies among the results, particularly a pronounced difference between allele frequency distributions called by VarScan and SomaticSniper. Survival analysis showed little robust predictive power for heterogeneity as measured by Mutant-Allele Tumor Heterogeneity (MATH) score, with the exception of uterine corpus endometrial carcinoma. However, we found that variations in mutant allele frequencies were mediated by variations in copy number. Our results indicate that the clinical predictions associated with MATH score are primarily caused by copy number aberrations that alter mutant allele frequencies. Finally, we present a mathematical model of linear tumor evolution demonstrating why MATH score is insufficient for distinguishing different scenarios of tumor growth. Our findings elucidate the importance of allele frequency distributions as a measure for tumor heterogeneity and their prognostic role.

The processes by which tumors evolve are essential to the efficacy of treatment, but quantitative understanding of intratumoral dynamics has been limited. Although intratumoral heterogeneity is common, quantification of evolution is difficult from clinical samples because treatment replicates cannot be performed and because matched serial samples are infrequently available. To circumvent these problems we derived and assayed large sets of human triple-negative breast cancer xenografts and cell cultures from two patients, including 86 xenografts from cyclophosphamide, doxorubicin, cisplatin, docetaxel, or vehicle treatment cohorts as well as 45 related cell cultures. We assayed these samples via exome-seq and/or high-resolution droplet digital PCR, allowing us to distinguish complex therapy-induced selection and drift processes among endogenous cancer subclones with cellularity uncertainty AUTHOR SUMMARYAn overarching challenge of cancer is that patients develop resistance to treatment -- an essentially evolutionary process. However, there is currently very little understanding of how tumor evolution can be exploited to improve treatment. One reason for this is that usually only 1-2 samples can be obtained per patient, so cancer evolutionary processes are still poorly understood. To solve this problem, we created many dozens of copies of the tumors from two breast cancer patients using xenografting and cell culture methods. We then compared the evolution in these tumor copies in response to different treatments, including four of the most common breast cancer chemotherapies. These studies present the most exhaustive comparisons of treatment-induced evolution that have yet been performed for individual cancer patients. Unexpectedly, high-resolution sequencing of these samples revealed a special dynamically treatable ecology in one tumor, in which tumor growth during platinum therapy was determined by the ecological balance of two tumor cell populations. Our work shows that ecologies that can be targeted by dynamic treatment strategies arise spontaneously in breast cancers. Population heterogeneity is common within cancers, and our work suggests how tracking of intratumoral evolution can be used to optimize treatment.

We read with great interest the paper by Williams, et. al.[1], who argued for neutral evolution in tumors by analyzing The Cancer Genome Atlas (TCGA) data. They supported this conclusion by showing high R2 values for fits to a neutral evolution model predicting M ∝ 1 / f, where M is the number of somatic mutations with allele frequency ≥ f. However, we believe a conclusion of neutrality must be treated cautiously, as high R2 values are consistent with many evolutionary models.

RNA-protein binding is critical to gene regulation, controlling fundamental processes including splicing, translation, localization and stability, and aberrant RNA-protein interactions are known to play a role in a wide variety of diseases. However, molecular understanding of RNA-protein interactions remains limited; in particular, identification of RNA motifs that bind proteins has long been challenging, especially when such motifs depend on both sequence and structure. Moreover, although RNA binding proteins (RBPs) often contain more than one binding domain, algorithms capable of identifying more than one binding motif simultaneously have not been developed. In this paper we present a novel pipeline to determine binding peaks in crosslinking immunoprecipitation (CLIP) data, to discover multiple possible RNA sequence/structure motifs among them, and to experimentally validate such motifs. At the core is a new semi-automatic algorithm SARNAclust, the first unsupervised method to identify and deconvolve multiple sequence/structure motifs simultaneously. SARNAclust computes similarity between sequence/structure objects using a graph kernel, providing the ability to isolate the impact of specific features through the bulge graph formalism. Application of SARNAclust to synthetic data shows its capability of clustering 5 motifs at once with a V-measure value of over 0.95, while GraphClust achieves only a V-measure of 0.083 and RNAcontext cannot detect any of the motifs. When applied to existing eCLIP sets, SARNAclust finds known motifs for SLBP and HNRNPC and novel motifs for several other RBPs such as AGGF1, AKAP8L and ILF3. We demonstrate an experimental validation protocol, a targeted Bind-n-Seq-like high-throughput sequencing approach that relies on RNA inverse folding for oligo pool design, that can validate the components within the SLBP motif. Finally, we use this protocol to experimentally interrogate the SARNAclust motif predictions for protein ILF3. Our results support a newly identified partially double-stranded UUUUUGAGA motif similar to that known for the splicing factor HNRNPC., Author summary RNA-protein binding is critical to gene regulation, and aberrant RNA-protein interactions play a role in a wide variety of diseases. However, molecular understanding of these interactions remains limited because of the difficulty of ascertaining the motifs that bind each protein. To address this challenge, we have developed a novel algorithm, SARNAclust, to computationally identify combined structure/sequence motifs from immunoprecipitation data. SARNAclust can deconvolve multiple motifs simultaneously and determine the importance of specific features through a graph kernel and bulge graph formalism. We have verified SARNAclust to be effective on synthetic motif data and also tested it on ENCODE eCLIP datasets, identifying known motifs and novel predictions. We have experimentally validated SARNAclust for two proteins, SLBP and ILF3, using RNA Bind-n-Seq measurements. Applying SARNAclust to ENCODE data provides new evidence for previously unknown regulatory interactions, notably splicing co-regulation by ILF3 and the splicing factor hnRNPC.

12050Background: CD73, known as ecto-5’-nucleotidase and encoded by the NT5E gene, is a pivotal enzyme that converts extracellular adenosine monophosphate (AMP) into adenosine, which promotes tumor...

Many conserved noncoding sequences function as transcriptional enhancers that regulate gene expression. Here, we report that protein-coding DNA also frequently contains enhancers functioning at the transcriptional level. We tested the enhancer activity of 31 protein-coding exons, which we chose based on strong sequence conservation between zebrafish and human, and occurrence in developmental genes, using a Tol2 transposable GFP reporter assay in zebrafish. For each exon we measured GFP expression in hundreds of embryos in 10 anatomies via a novel system that implements the voice-recognition capabilities of a cellular phone. We find that 24/31 (77%) exons drive GFP expression compared to a minimal promoter control, and 14/24 are anatomy-specific (expression in four anatomies or less). GFP expression driven by these coding enhancers frequently overlaps the anatomies where the host gene is expressed (60%), suggesting self-regulation. Highly conserved coding sequences and highly conserved noncoding sequences do not significantly differ in enhancer activity (coding: 24/31 vs. noncoding: 105/147) or tissue-specificity (coding: 14/24 vs. noncoding: 50/105). Furthermore, coding and noncoding enhancers display similar levels of the enhancer-related histone modification H3K4me1 (coding: 9/24 vs noncoding: 34/81). Meanwhile, coding enhancers are over three times as likely to contain an H3K4me1 mark as other exons of the host gene. Our work suggests that developmental transcriptional enhancers do not discriminate between coding and noncoding DNA and reveals widespread dual functions in protein-coding DNA.

Background The patterns of mutation vary both within and across genomes. It has been shown for a few mammals that mutation rates vary within the genome, while for unknown reasons, the sensu stricto yeasts have uniform rates instead. The generality of these observations has been unknown. Here we examine silent site substitutions in a more expansive set (20 mammals, 27 fungi, 4 insects) to determine why some genomes demonstrate this mosaic distribution and why others are uniform. Results We applied several intragene and intergene correlation tests to measure regional substitution patterns. Assuming that silent sites are a reasonable approximation to neutrally mutating sequence, our results show that all multicellular eukaryotes exhibit mutational heterogeneity. In striking contrast, all fungi are mutationally uniform – with the exception of three Candida species: C. albicans, C. dubliniensis, and C. tropicalis. We speculate that aspects of replication timing may be responsible for distinguishing these species. Our analysis also reveals classes of genes whose silent sites behave anomalously with respect to the mutational background in many species, indicating prevalent selective pressures. Genes associated with nucleotide binding or gene regulation have consistently low silent substitution rates in every mammalian species, as well as multiple fungi. On the other hand, receptor genes repeatedly exhibit high silent substitution rates, suggesting they have been influenced by diversifying selection. Conclusion Our findings provide a framework for understanding the regional mutational properties of eukaryotes, revealing a sharp difference between fungi and multicellular species. They also elucidate common selective pressures acting on eukaryotic silent sites, with frequent evidence for both purifying and diversifying selection.

COMIT is presented; an algorithm for detecting functional non-coding motifs in coding regions, separating nucleotide and amino acid effects., Coding nucleotide sequences contain myriad functions independent of their encoded protein sequences. We present the COMIT algorithm to detect functional noncoding motifs in coding regions using sequence conservation, explicitly separating nucleotide from amino acid effects. COMIT concurs with diverse experimental datasets, including splicing enhancers, silencers, replication motifs, and microRNA targets, and predicts many novel functional motifs. Intriguingly, COMIT scores are well-correlated to scores uncalibrated for amino acids, suggesting that nucleotide motifs often override peptide-level constraints.

Background The rate at which neutral (non-functional) bases undergo substitution is highly dependent on their location within a genome. However, it is not clear how fast these location-dependent rates change, or to what extent the substitution rate patterns are conserved between lineages. To address this question, which is critical not only for understanding the substitution process but also for evaluating phylogenetic footprinting algorithms, we examine ancestral repeats: a predominantly neutral dataset with a significantly higher genomic density than other datasets commonly used to study substitution rate variation. Using this repeat data, we measure the extent to which orthologous ancestral repeat sequences exhibit similar substitution patterns in separate mammalian lineages, allowing us to ascertain how well local substitution rates have been preserved across species. Results We calculated substitution rates for each ancestral repeat in each of three independent mammalian lineages (primate – from human/macaque alignments, rodent – from mouse/rat alignments, and laurasiatheria – from dog/cow alignments). We then measured the correlation of local substitution rates among these lineages. Overall we found the correlations between lineages to be statistically significant, but too weak to have much predictive power (r 2 <5%). These correlations were found to be primarily driven by regional effects at the scale of several hundred kb or larger. A few repeat classes (e.g. 7SK, Charlie8, and MER121) also exhibited stronger conservation of rate patterns, likely due to the effect of repeat-specific purifying selection. These classes should be excluded when estimating local neutral substitution rates. Conclusion Although local neutral substitution rates have some correlations among mammalian species, these correlations have little predictive power on the scale of individual repeats. This indicates that local substitution rates have changed significantly among the lineages we have studied, and are likely to have changed even more for more diverged lineages. The correlations that do persist are too weak to be responsible for many of the highly conserved elements found by phylogenetic footprinting algorithms, leading us to conclude that such elements must be conserved due to selective forces.

Background It has been increasingly appreciated that coding sequences harbor regulatory sequence motifs in addition to encoding for protein. These sequence motifs are expected to be overrepresented in nucleotide sequences bound by a common protein or small RNA. However, detecting overrepresented motifs has been difficult because of interference by constraints at the protein level. Sampling-based approaches to solve this problem based on codon-shuffling have been limited to exploring only an infinitesimal fraction of the sequence space and by their use of parametric approximations. Results We present a novel O(N(log N)2)-time algorithm, CodingMotif, to identify nucleotide-level motifs of unusual copy number in protein-coding regions. Using a new dynamic programming algorithm we are able to exhaustively calculate the distribution of the number of occurrences of a motif over all possible coding sequences that encode the same amino acid sequence, given a background model for codon usage and dinucleotide biases. Our method takes advantage of the sparseness of loci where a given motif can occur, greatly speeding up the required convolution calculations. Knowledge of the distribution allows one to assess the exact non-parametric p-value of whether a given motif is over- or under- represented. We demonstrate that our method identifies known functional motifs more accurately than sampling and parametric-based approaches in a variety of coding datasets of various size, including ChIP-seq data for the transcription factors NRSF and GABP. Conclusions CodingMotif provides a theoretically and empirically-demonstrated advance for the detection of motifs overrepresented in coding sequences. We expect CodingMotif to be useful for identifying motifs in functional genomic datasets such as DNA-protein binding, RNA-protein binding, or microRNA-RNA binding within coding regions. A software implementation is available at http://bioinformatics.bc.edu/chuanglab/codingmotif.tar