Trey Ideker Research Articles

Abstract 2348: A global multiscale map of protein assemblies from integration of protein interactions and images

Abstract Cells regulate growth and function through a hierarchical structure of subcellular protein assemblies, in which alterations can result in cellular dysfunction leading to disease such as cancer. Much of this structure remains uncharted, resulting in recent efforts to map subcellular organization at different physical scales. Here, we report a global architectural map of human cancer cell protein assemblies spanning 10−9 to 10−5 nm, based on integration of near-proteome-wide affinity purification mass spectrometry-based protein interactions and immunofluorescent imaging in U-2 OS osteosarcoma cancer cells. The U-2 OS multi-scale integrated cell map places >5000 proteins into 270 distinct subcellular protein assemblies across biological scales, representing approximately half of expressed proteins. There are known organelles and protein complexes recovered in the map, as well as 152 putative assemblies, such as a putative interferon signaling complex consisting of a serine protease and STAT transcription factors. The map also incorporates 128 previously uncharacterized proteins into protein assemblies, such as C18orf21 association with a canonical RNAse complex. Protein subsystems in the U-2 OS map were evaluated for the potential for downstream integrative structural modeling, where available structural data (e.g. crosslinking, Protein Data Bank structures, AlphaFold predictions) are combined in a Bayesian framework to model the physical structures of protein complexes. Using this framework, we created an integrated structural model of new proteins interacting with the Rag-Ragulator complex, which regulates MAPK and mTOR signaling. In summary, the global proteome architecture map provides a resource for cellular biology discovery and the systematic determination of protein functions and structures. Citation Format: Leah V. Schaffer, Mengzhou Hu, Edward L. Huttlin, Gege Qian, Abantika Pal, Neelesh Soni, Andrew Latham, Laura Pontano Vaites, Trang Le, Yue Qin, Christopher Churas, Dexter Pratt, Ignacia Echeverria, Andrej Sali, J Wade Harper, Steven P. Gygi, Emma Lundberg, Trey Ideker. A global multiscale map of protein assemblies from integration of protein interactions and images [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2348.

Abstract 3287: Learning to target CDK4/6 inhibitor resistance via a breast cancer-specific atlas of cellular mechanisms

Abstract Development of resistance to targeted, chemotherapeutic, and immune-oncology treatments alike is a major barrier to long-term remission in advanced cancer patients. We developed an integrated computational-experimental platform to mitigate treatment resistance by identifying therapeutics that selectively target clinical resistance mechanisms. The key innovation is the computational framework that comprehensively maps tumor-specific mechanisms of response and resistance to a treatment based on clinicogenomic and/or preclinical pharmacogenomic data. It then screens for drugs or targets that are synthetically lethal with these resistance mechanisms. Here we present the identification of therapeutic options that mitigate resistance to CDK4/6 inhibitors as a case study. Leveraging a breast cancer-specific atlas of cellular architecture, our framework learned genetic mechanisms of CDK4/6 inhibitor resistance from a high-throughput phenotypic screen of ~700 cell lines treated with the clinical CDK4/6 inhibitor palbociclib. The learned mechanisms formed a parsimonious set of hierarchically linked protein complexes that coordinate G1-to-S transition (P=5.6 × 10−12). The mechanisms unified distinct resistant populations such as samples resistant due to CDK4/6-Rb checkpoint bypass and those resistant due to growth factor receptor activation. Applied to 70 ER+ metastatic breast cancer patients in a real-world dataset, the learned resistance mechanisms accurately predicted patient response to palbociclib (P=3 × 10−4, log-rank test). Computationally screening 20K clinical-grade small molecules and gene targets for therapeutic options synthetically lethal with the learned resistance mechanisms identified the known CDK4/6 resistance targets CDK2, CCNE1, and E2F. It also identified a target that, to our knowledge, has not previously been rigorously evaluated for its potential to mitigate CDK4/6 inhibitor resistance. In-vitro, monotherapy inhibition of the target by siRNA selectively decreased growth of breast cancer cell lines resistant to CDK4/6 inhibitors by 52% compared to CDK4/6i sensitive cell lines (P=0.023, Mann-Whitney U-test). As a combination therapy, inhibition of the target by siRNA selectively increased the sensitivity to palbociclib by nearly an order of magnitude in an otherwise resistant breast cancer cell line (expected vs observed IC50: 9.7 uM vs. 1.3 uM) while not having a substantial effect in a palbociclib-sensitive cell line (expected vs observed IC50: 0.57 uM vs. 0.56 uM). Finally, while monotherapy inhibition of the target and CDK4/6 induced cell cycle arrest, the combination strongly induced apoptosis in cell lines resistant to CDK4/6 inhibitors. These results highlight how computational deconvolution of tumor resistance mechanisms enable algorithmic identification of therapeutic options to target treatment-resistant cell populations. Citation Format: Adam Yaari, Eduardo Farias, Francisco Guedes, Oliver Priebe, Lee McDaniel, Tyler Earnest, Trey Ideker, Maxwell A. Sherman. Learning to target CDK4/6 inhibitor resistance via a breast cancer-specific atlas of cellular mechanisms [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3287.

Abstract 6543: A map of osteosarcoma cell architecture reveals convergence of pediatric cancer mutations on protein assemblies

Abstract Pediatric tumors are characterized by a low incidence of somatic mutations, which occur in different genomic patterns from adult cancers. For these reasons, interpreting the functional impact of mutations in pediatric cancer patients has remained difficult. One approach to interpreting rare coding mutations is to analyze the convergence of these mutations on larger structural and functional units, such as protein complexes, super assemblies, biomolecular condensates, and organelles. Towards this aim, we created a cancer cell map in the osteosarcoma cell line based on proteome-wide integration of protein interactions and immunofluorescent imaging in U-2 OS pediatric bone cancer cells. This multiscale integrated cell map of U-2 OS cell organizes 5,254 proteins into 270 protein assemblies. Using this multiscale cell map, we analyze 772 genomes from 18 cancer types reported in the DKFZ pan-pediatric cancer study to identify 25 protein assemblies in the map under mutational pressure. This analysis reveals mutated protein assemblies involved with chromatin maintenance and remodeling as well as cell integrity and mobility. For example, we identify a Collagen Biosynthesis and Remodeling Complex, as mutated in 9 cancer types and 15 patients in the dataset, indicating extracellular matrix remodeling is associated with pediatric cancer progression. We also identify a putative biological condensate, which we call the Nuclear Transcription Suppression Complex (NTSC). This protein assembly is found mutated in 20 patients across 10 cancer types, implicating intrinsically disordered proteins in development of pediatric cancer. Overall, this work illustrates how a global multiscale map of cell architecture enables a mechanistic understanding of clinical patient genomes. Citation Format: Mengzhou Hu, Leah V. Schaffer, Edward L. Huttlin, Gege Qian, Andrew Latham, Abantika Pal, Laura Pontano Vaites, Trang Le, Yue Qin, Christopher Churas, Dexter Pratt, Robin Bachelder, Peter Zage, Ignacia Echeverria, Andrej Šali, J. Wade Harper, Steven P. Gygi, Emma Lundberg, Trey Ideker. A map of osteosarcoma cell architecture reveals convergence of pediatric cancer mutations on protein assemblies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6543.

Abstract 7101: Remodeling of cancer cells by chemotherapy

Abstract Cisplatin is a commonly administrated chemotherapy drug for cancer treatment. Although the direct mechanism of cisplatin is largely well-defined, it is challenging to comprehensively assess its systemic effects. Cellular responses to perturbations are intricate processes involving modifications to protein complexes across diverse cellular compartments. Therefore, substantial research endeavors have been directed towards unraveling protein organization in cells. We recently built a multiscale integrated cell map, MuSIC, that combines fluorescence-stained images and interactions derived from affinity purification mass spectrometry to allow comprehensive characterization of cellular structure and functional networks of 5,254 proteins of the human bone osteosarcoma epithelial line (U2-OS). To exploit the biology of U2OS, we integrated dynamics into this map to determine the proteome reorganization upon cisplatin exposure. Thus, we developed a data-driven approach integrating high throughput size exclusion chromatography mass spectrometry (SEC-MS) data with large-scale cell maps built from static-state information to systematically map at all scales the remodeling of cell architecture in drug-treated cells. Here, we acquired drug-perturbed protein-protein interactions (PPIs) from the SEC-MS data by treating U2-OS with cisplatin. Integrating the differential interactions identified from SEC-MS PPIs with the scaffold large-scale hierarchy (MuSIC) allows retention of the upper cellular organization to systematically map drug impact on protein complexes in the cell. This pipeline serves as an explorative method to identify the drug-responsive organelles, alterations of hierarchical structures, and translocation of proteins. Apart from the well known alterations of DNA damage repair complexes, we also recapitulate remodeling in metabolism pathways, splicing, cAMP-dependent kinase activity, and histone modification. We identify 26 novel protein assemblies containing members that are previously associated with cisplatin biology. This pipeline exploits the potential of a well-established network by integrating the high throughput data modality, allowing for the screening of a range of chemical agents and stimuli to generate differential networks. Citation Format: Gege Qian, Leah Schaffer, Kyung-Mee Moon, Leonard Foster, Trey Ideker, Mengzhou Hu, Sahar Alkhairy, Emma Lundberg. Remodeling of cancer cells by chemotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7101.

Abstract 4899: Interface-guided phenotyping of coding variants in the transcription factor RUNX1

Abstract Understanding the consequences of single amino acid substitutions in cancer driver genes remains an unmet need. Perturb-seq provides a tool to investigate the effects of individual mutations on cellular programs. Here we deploy SEUSS, a Perturb-seq like approach, to generate and assay mutations at physical interfaces of the RUNX1 Runt domain. We measured the impact of 115 mutations on RNA profiles in single myelogenous leukemia cells and used the profiles to categorize mutations into three functionally distinct groups: wild-type (WT)-like, loss-of-function (LOF)-like and hypomorphic. Notably, the largest concentration of functional mutations (non-WT-like) clustered at the DNA binding site and contained many of the more frequently observed mutations in human cancers. Hypomorphic variants shared characteristics with loss of function variants but had gene expression profiles indicative of response to neural growth factor and cytokine recruitment of neutrophils. Additionally, DNA accessibility changes upon perturbations were enriched for RUNX1 binding motifs, particularly near differentially expressed genes. Overall, our work demonstrates the potential of targeting protein interaction interfaces to better define the landscape of prospective phenotypes reachable by amino acid substitutions. Citation Format: Kivilcim Ozturk, Rebecca Panwala, Jeanna Sheen, Kyle Ford, Nathan Jayne, Dong-Er Zhang, Stephan Hutter, Torsten Haferlach, Trey Ideker, Prashant Mali, Hannah Carter. Interface-guided phenotyping of coding variants in the transcription factor RUNX1 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4899.

Abstract 4918: Cancer mutations converge on a constellation of molecular assemblies to predict resistance to replication stress agents

Abstract Rapid proliferation is a hallmark of tumor cells, associated with sensitivity to therapeutics that cause DNA replication stress (RS). Many tumors exhibit drug resistance, however, via molecular pathways that are incompletely understood. Here, we develop an ensemble of predictive models that elucidate how cancer mutations impact the response to common RS-inducing (RSi) agents. The models implement recent advances in deep learning to facilitate multi-drug prediction and mechanistic interpretation. Instead of associating individual genetic alterations with RSi drug responses directly, the strategy is to project these alterations on a map of protein complexes and larger molecular assemblies associated with cancer. Initial studies in tumor cells identify 41 molecular assemblies that integrate alterations in hundreds of genes for accurate drug response prediction. These cover roles in transcription, repair, cell-cycle checkpoints, and growth signaling, of which 30 are shown by loss-of-function genetic screens to regulate drug sensitivity or replication restart. The model translates to cisplatin-treated cervical cancer patients, highlighting an RTK-JAK-STAT assembly governing resistance. This study defines a compendium of mechanisms by which mutations affect therapeutic responses, with implications for drug selection and combination. Citation Format: Xiaoyu Zhao, Akshat Singhal, SungJoon Park, JungHo Kong, Robin Bachelder, Trey Ideker. Cancer mutations converge on a constellation of molecular assemblies to predict resistance to replication stress agents [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4918.

Abstract 7383: G2PT: Mechanistic genotype-phenotype translation using hierarchical transformers

Abstract Motivation: Many machine learning models have been proposed that use genotypes to predict various phenotypes. Recently, these models have focused not only on an accurate prediction but mechanistic interpretation, in that they attempt to describe the hierarchy of biological systems underlying the predicted phenotype. Such models still face major challenges, however, including how to robustly quantify the importance of systems mediating phenotypic outcomes and how to represent the bidirectional flow of information among genes, systems, phenotypes, and cell environment. Results: To address these challenges we introduce G2PT, a general Genotype-to-Phenotype Transformer, which we apply here to the classic G2P problem of modeling tumor cell sensitivity to drugs. We show that the framework of hierarchical transformers can represent biological systems spanning a wide range of physical scales and provide a quantitative assessment of the interplay between biological components important for phenotype prediction. When used to analyze a high-throughput cell screen of 436 drugs exposed to 1,097 cell types, we show that the model achieves good performance in predicting cell sensitivity for 83% of drugs (⍴ > 0.5 in cross-validation). G2PT provides a robust and interpretable model with potential application to many challenges in genotype-phenotype translation. Citation Format: Ingoo Lee, Sungjoon Park, Hojung Nam, Trey Ideker. G2PT: Mechanistic genotype-phenotype translation using hierarchical transformers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7383.

Abstract A015: Prediction of immunotherapy response using mutations to cancer protein assemblies

Abstract While immune checkpoint inhibitors (ICI) have substantially advanced cancer treatment, favorable responses are limited to a subset of patients and are not strongly predicted by histology or overall mutation burden. Here, we factor mutational burden across a landscape of known and putative multimeric protein assemblies, yielding Assembly Mutation Burden (AMB) scores for each patient. Predictive modeling identifies 14 assemblies in which mutation is associated with ICI response in 555 bladder cancer (BLCA) and non-small-cell lung cancer (NSCLC) patients, outperforming predictions based on genome-wide mutation burden alone. The assemblies capture immunogenic hallmarks and accurately predict treatment outcomes in a validation cohort of 68 immunotherapy-treated NSCLC patients. We explore a predictive assembly of 50 immunoregulatory complex proteins, in which we show CRISPR/Cas9 genetic disruptions influence ICI-treatment outcomes in mice. This study provides a roadmap for using knowledge of tumor cell biology to interpret the immunogenome. Citation Format: JungHo Kong, Robin Bachelder, Xiaoyu Zhao, Sungjoon Park, Akshat Singhal, Hannah Carter, Trey Ideker. Prediction of immunotherapy response using mutations to cancer protein assemblies [abstract]. In: Proceedings of the AACR-NCI-EORTC Virtual International Conference on Molecular Targets and Cancer Therapeutics; 2023 Oct 11-15; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2023;22(12 Suppl):Abstract nr A015.

Abstract 4306: Identifying genetic dependencies of replication stress using interpretable artificial intelligence

Abstract Rapid proliferation is a hallmark of tumor cells, associated with sensitivity to chemicals that cause DNA replication stress (RS). Due to sustained proliferative signaling and/or defective DNA repair, cancer cells undergo persistent RS making them strongly dependent on the replication stress response (RSR). A consequence of this dependency is that RS becomes an exploitable therapeutic vulnerability in cancer treatment. Classical RS drugs work predominantly by interfering with DNA replication in dividing cells. Recently, an increasing number of drugs have been designed to specifically target RSR proteins. However, molecular pathways responsible for drug response are incompletely understood. Here we build an interpretable deep-learning model aimed at understanding mechanisms of susceptibility and resistance to replicative stress. Instead of associating genetic alterations with drug responses directly, our approach is to project individual mutations on a map of protein complexes and larger molecular assemblies with prior evidence for involvement in cancer. This approach is prompted and supported by the concept that cancer is a network-based disease arising from the action of hallmark cancer pathways. Through systematic interpretation, the model identifies 37 complexes that integrate rare alterations in hundreds of genes for accurate response prediction. The complexes, which cover roles in transcription, DNA repair, cell-cycle checkpoint, and immunity, are further investigated by directed genetic perturbations, validating 24 for which RS effects are phenocopied by CRISPR guide RNAs. For complexes with poorly characterized functions, further insights are obtained via their profiles of in-silico activation across RS agents. This study creates a library of system-level genetic vulnerabilities governing replication stress, with implications for drug selection and combination. Citation Format: Xiaoyu Zhao, Akshat Singhal, Trey Ideker. Identifying genetic dependencies of replication stress using interpretable artificial intelligence. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 4306.

Abstract PO017: HBV alters YAP regulation in liver cancer by remodeling PP2A complexes

Abstract Viral hepatitis promotes hepatocellular carcinoma (HCC) initiation by inducing inflammation, cellular stress and cell death. However, the cell-intrinsic effects of viral infection in advanced HCC remain unclear. We used affinity purification mass spectrometry to develop a complete map of 145 interactions among Hepatitis B Virus (HBV) and host proteins in the HUH7 HCC tumor cell line, identifying known and novel HBV/host protein-protein interactions. We integrated this map with HCC genomes, identifying 61 proteins with preferential mutation in non-HBV HCC, suggesting that their interaction with HBV plays a role in cancer. Focusing on proteins that directly interact with the HBV oncoprotein X (HBx), we found that HBx rewires the effect of the PP2A phosphatase on HCC signaling. HBx binding excluded striatin-family regulatory subunits from the PP2A complex, causing Hippo kinase activation. This effect creates a requirement for integrin signaling to mTOR complex 2 to maintain expression of the YAP oncoprotein, critical for HCC growth. Thus, HBV rewires HCC signaling and may promote targetable dependencies. Citation Format: John Gordan, Adriana Pitea, Rigney E Turnham, Manon Eckhardt, Gwendolyn M Jang, Huat Lim, Alex L Choi, Sourav Bandyopadhyay, Danielle Swaney, Kevan Shokat, Trey Ideker, Nevan Krogan. HBV alters YAP regulation in liver cancer by remodeling PP2A complexes [abstract]. In: Proceedings of the AACR Special Conference: Advances in the Pathogenesis and Molecular Therapies of Liver Cancer; 2022 May 5-8; Boston, MA. Philadelphia (PA): AACR; Clin Cancer Res 2022;28(17_Suppl):Abstract nr PO017.

Abstract 636: Prediction of therapeutic response via data-driven maps of tumor cell architecture

Abstract Machine learning is revolutionizing the analysis of biological and biomedical data. One continued challenge is to build models that are not only predictive but also provide mechanistic explanations. Directly incorporating explainability into the model’s design has emerged as a possible solution. Recently, we reported several “visible” machine learning systems - DCell and DrugCell - which predict cellular growth based on genetic perturbations. In these visible models, the structure of the underlying neural network mirrors known aspects of cell biology. Here we designed a cancer-specific model of therapeutic response. First, we selected NeST (Nested Systems in Tumors; Zheng et al., Science 2021), a data-driven hierarchy of tumor cell systems under selection in cancer, as the structure for our visible neural network. Second, we restricted input features to genes currently assayed on clinical cancer gene panels. We considered training schemes that assess the model’s ability to generalize to novel genotypes. Most importantly, we assessed performance on patient data from the AACR Project GENIE clinical trial where we discriminated between patients who are likely to respond to selected drug treatments. Finally, we examined model explanations within the visible NeST architecture to highlight cellular mechanisms of response to drug therapy. Overall, we found that by considering a tumor cell architecture optimized for clinical application, we can design an interpretable deep learning model that both accurately stratifies patient responses and explains the drug mechanism in cancer cells. Citation Format: Erica N. Silva, Akshat Singhal, Sungjoon Park, Jason Kreisberg, Trey Ideker. Prediction of therapeutic response via data-driven maps of tumor cell architecture [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 636.

Abstract 1159: Predicting clinical drug responses using a few-shot learning-based interpretable AI

Abstract High-throughput screens have generated large amounts of data characterizing how thousands of cell lines respond to hundreds of anti-cancer therapies. However, predictive drug response models trained using data from cell lines often fail to translate to clinical applications. Here, we focus on two key issues to improve clinical performance: 1) Transferability: the ability of predictive models to quickly adapt to clinical contexts even with a limited number of samples from patients, and 2) Interpretability: the ability to explain how drug-response predictions are being made given an individual patient’s genotype. Notably, an interpretable AI model can also help to identify biomarkers of treatment response in individual patients. By leveraging new developments in meta-learning and interpretable AI, we have developed an interpretable drug response prediction model that is trained on large amounts of data from experiments using cell lines and then transferred to clinical applications. We assessed our model’s clinical utility using AACR Project GENIE data, which contains mutational profiles from tumors and the patient’s therapeutic responses. We have demonstrated the feasibility of applying our AI-driven predictive model to clinical settings and shown how this model can support clinical decision making for tumor boards. Citation Format: Sungjoon Park, Akshat Singhal, Erica Silva, Jason F. Kreisberg, Trey Ideker. Predicting clinical drug responses using a few-shot learning-based interpretable AI [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 1159.

Abstract 217: Mutations in iron-sulfur cluster proteins are associated with decreased bladder cancer cell senescence and improved survival outcomes

Abstract Recent tumor genome analyses have cataloged hundreds of mutated genes, including a large population of rarely mutated genes with ill-defined roles in cancer. Here we find that, when considering the 10 genes of the iron-sulfur (Fe-S) pathway as an integrated whole, somatic mutations occur with a higher than expected frequency in 21.3% of bladder tumors. Mutations in these genes are associated with increased survival and mutational burden. Analysis of tumor gene expression highlights a relationship between these Fe-S mutations and signalling regulating proliferation and quiescence, with mutations promoting proliferative programs and abolishing maintenance of quiescence. In bladder tumor cells, we show that replication of this proliferative program, using an DYRK1A kinase inhibitor, promotes sensitivity to genotoxic agents. Furthermore, disruption of Fe-S genes ERCC2, MMS19, CIOA1, BRIP1 or POLD1 promotes cell proliferation and concomitant sensitivity to genotoxic agents. These results suggest a model in which Fe-S pathway mutations are selected due to their ability to bypass DNA damage checkpoints and drive proliferation, an aspect which simultaneously creates vulnerability to chemotherapy and resultant improvements in survival. Citation Format: Devin Patel, Fan Zheng, Michael Chen, Jason Kreisberg, Erica Silva, Trey Ideker, Kate Licon. Mutations in iron-sulfur cluster proteins are associated with decreased bladder cancer cell senescence and improved survival outcomes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 217.

Abstract 178: Desmosome mutations in melanoma promote cellular proliferation and disease progression

Abstract Desmosomes are transmembrane protein complexes that contribute to cell-cell adhesion in the epithelia and other tissues under mechanical stress. Aberrant desmosome expression is often associated with developmental diseases leading to impaired tissue integrity. Recently, similar findings have been reported in cancer; Mutations in desmosomes genes have been observed in various cancer types including skin cancer, head and neck and lung cancer, however mostly epigenetic alterations have been used to associate desmosomes as suppressors of tumor metastasis. Here, we report that desmosomes are frequently mutated in seven cancer types. In melanoma, we find that over 70% of tumors have non-synonymous mutations in desmosomes, and that the desmosome mutational burden is associated with a strong decrease in mRNA expression levels in primary tumor samples (R = -0.23). Differential gene expression analysis and functional characterizations between mutant and wild-type tumors implicates the mutated cells in promoting cell proliferation at early stages of tumorigenesis. These results emerge uniquely from a systems-level analysis integrating multiple proteins in complexes and multiple cell types in heterogeneous tumors. Citation Format: Maayan Baron, Trey Ideker. Desmosome mutations in melanoma promote cellular proliferation and disease progression [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 178.

Abstract 2133: Systematic mapping of genetic interactions in human cancer cells reveals context dependent cancer signaling pathways

Abstract Genetic interaction screens have long been used to map gene functions and cellular pathways in model organisms and more recently in human cells. However, due to the technical challenges of these maps, most studies in human cells have focused on a single cell line, limiting the ability to determine how different genetic and cellular contexts influence genetic interactions. Here, we engineered a 110,728-element combinatorial CRISPR-Cas9 library to systematically map 12,282 genetic interactions among tumor suppressor genes and druggable targets. We mapped the landscape of genetic interactions in cell lines from breast cancer, head and neck squamous carcinoma, and non-small cell lung cancer as well as in non-cancerous, epithelial cells. The resulting genetic interaction maps emphasize the context dependency of these interactions. Of the interactions that are conserved across multiple cell lines, many include frequently essential genes, suggesting that these interactions may be critical for fundamental cellular functions. We found that genetic interaction maps from cancer cell lines are more similar to one another than non-cancer interaction maps, even between cancer and normal cells from the same tissue. Subsequent targeted chemogenetic screens confirmed interaction hubs across multiple cell lines while highlighting the robustness of some of the strongest interactions. Many of the identified genetic interactions suggest therapeutic interventions, such as the combination of CDK4 and MAPK inhibitors. Using the tissue-specific genetic interaction networks, we also built a deep learning model that accurately predicts patient outcomes, outperforming models that use accepted biomarkers such as MammaPrint. In summary, these genetic interaction maps provide a broad resource to investigate context dependency in cancer signaling pathways and in cancer drug response. Citation Format: Samson H. Fong, Brent M. Kuenzi, John Lee, Kyle Sanchez, Ana Bojorquez-Gomez, Kyle Ford, Brenton P. Munson, Katherine Licon, Jeff Hager, John Paul Shen, Jason F. Kreisberg, Prashant Mali, Trey Ideker. Systematic mapping of genetic interactions in human cancer cells reveals context dependent cancer signaling pathways [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2133.

Abstract 2078: A multiscale map of DNA damage response in human cells

Abstract The DNA damage response (DDR) ensures error-free genome replication and transcription, but is often disrupted in cancer. We developed DDRAM, a multilayered map of DDR derived from our own high-throughput physical interaction screen as well as genome, transcriptome, and proteome datasets from published sources. DDRAM organizes 328 genes into 41 hierarchical pathways. We identify 55 (17%) novel DDR genes with functions distributed across various pathways, and novel and more specific roles for 153 (47%) of established DDR genes. We perform epistasis mapping to support novel pathway assignments of FOXK1 in nucleotide metabolism and XRCC3 in the Fanconi anemia pathway. Using quantitative imaging of a fluorescent probe, in some cases confirmed by epistasis mapping, we experimentally validate members of the Short Patch Base Excision Repair pathway over a time course, revealing a more complete model of poly(ADP-ribose)-dependent recruitment to sites of DNA damage. Citation Format: Anton Kratz, Minkyu Kim, Fan Zheng, Christopher A. Koczor, Samson Fong, Jianfeng Li, Robert W. Sobol, Nevan Krogan, Trey Ideker. A multiscale map of DNA damage response in human cells [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2078.

Abstract 2308: The protein interaction landscape of breast cancer

Abstract Cancers have been associated with a diverse array of genomic alterations, many of which are rare with unknown significance. To understand the cellular mechanisms impacted by such alterations in breast invasive carcinoma, we have applied affinity-purification mass spectrometry to delineate comprehensive biophysical interaction networks for 40 frequently altered breast cancer proteins across three human breast cell lines, providing a novel resource of context-specific and shared protein-protein interaction networks in breast cancer cells. These networks interconnect and enrich for common and rare cancer mutations, and are substantially rewired by mutations. Our analysis identifies novel PIK3CA-interacting proteins which repress AKT signaling, and UBE2N emerges as a BRCA1 interactor predictive of clinical response to PARP inhibition. We also show that Spinophilin interacts with and dephosphorylates BRCA1 to promote DNA double-strand break repair. Thus, cancer protein interaction landscapes provide a framework for recognizing oncogenic drivers and drug vulnerabilities. Citation Format: Minkyu Kim, Jisoo Park, Mehdi Bouhaddou, Kyumin Kim, Ajda Rojc, Maya Modak, Margaret Soucheray, Patrick O'Leary, Denise Wolf, Dominique C. Mitchell, Fan Zheng, John D. Gordan, Jean-Philippe Coppé, Danielle L. Swaney, Laura van' t Veer, Alan Ashworth, Trey Ideker, Nevan J. Krogan. The protein interaction landscape of breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2308.

Abstract IA-09: Mapping cell structure across scales by fusing protein images and interactions

Abstract Cancer cells are multi-scale structures with modular organization across at least four orders of magnitude. Two central approaches for mapping this structure – protein fluorescent imaging and protein interaction – each generate extensive datasets but of distinct qualities and resolutions that are typically treated separately. Here, I will describe ongoing efforts to reconcile the large-scale data emerging from protein imaging and protein affinity purification interaction screens to create unified maps of cancer cells. The core concept is to use machine learning to compute a suitable embedding for each dataset, then to use these data embeddings to produce a general measure of protein distance calibrated against cellular components of known size. The evolving map, called the Multi-Scale Integrated Cell (MuSIC 1.0), currently resolves 69 subcellular systems of which approximately half are undocumented. Based on these findings we have performed 134 additional affinity purifications, validating close subunit associations for the majority of systems. I will also discuss how these cell maps can be used to identify protein complexes, and larger-scale systems such as organelles, that are under significant selective pressure for somatic mutations in different tumor types. In many of these cases, the mutation burden of the complex is far more substantial than the mutation frequency of any of the individual genes that encode it. By integration across scales, MuSIC substantially increases the mapping resolution obtained from imaging while giving protein interactions a spatial dimension, paving the way to incorporate many molecular data types in proteome-wide maps of cells. This work is a collaboration between the Cancer Cell Mapping Initiative (CCMI), the Human Protein Atlas (HPA), and the BioPlex Resource. Citation Format: Trey Ideker. Mapping cell structure across scales by fusing protein images and interactions [abstract]. In: Proceedings of the AACR Virtual Special Conference on Artificial Intelligence, Diagnosis, and Imaging; 2021 Jan 13-14. Philadelphia (PA): AACR; Clin Cancer Res 2021;27(5_Suppl):Abstract nr IA-09.

Abstract 4891: Hepatitis B virus remodels host protein interaction networks to generate distinct cellular dependencies

Abstract Hepatocellular carcinoma (HCC) is the second leading cause of cancer death worldwide. Advanced HCC has proven particularly difficult to treat because of a scarcity of clear genetic drivers of cancer progression; thus, there are currently no predictive markers that guide HCC therapy. HCC arises in the context of co-morbid hepatitis due to hepatitis B virus (HBV), hepatitis C (HCV) or fatty liver disease. We hypothesize that protein-protein interactions (PPIs) between viral proteins and HCC genes may contribute to tumor initiation and maintenance. In order to characterize these PPIs, we performed affinity purification - mass spectrometry (APMS), defining 145 HBV/host PPIs including known and novel interacting partners. We next used a network propagation algorithm to identify host genes and protein complexes that were preferentially mutated in the absence of HBV infection. HBV is a small DNA virus, with 4 genes of which only one has enzymatic activity, raising a question as to how HBV interaction modifies host behavior. Using AP-MS of host proteins, we found that the HBV X protein (HBx) remodels multiple host protein complexes through direct interaction. These physical effects on complex components result in distinct biochemical behavior from the CRL4 E3 ubiquitin ligase complex as well as the phosphatase PP2A, as determined through global phosphoproteomics and ubiquitin analysis. We show that this remodeling driven by HBx substantially changes cellular protein turnover and downstream signaling dynamics. We followed this up with assessments of cellular viability and proliferation in response to pharmacological inhibition or CRISPRi-based knockdown of HBx effectors. Our data support a model where HBV proteins alter the components and behavior of key regulatory protein complexes in the cell, altering tumor behavior and raising the possibility of precision therapeutics for HCC. Citation Format: John D. Gordan, Adriana Pitea, Manon Eckhardt, Gwendolyn Jang, Rigney E. Turnham, Alex L M. Choi, John Von Dollen, Huat C. Lim, Elizabeth F. Thayer, R. Katie Kelley, Danielle L. Swaney, Wei Zhang, Fabian J. Theis, Trey Ideker, Nevan J. Krogan. Hepatitis B virus remodels host protein interaction networks to generate distinct cellular dependencies [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 4891.

Abstract 223: Using physical, genetic and integrated cancer cell maps to investigate head and neck cancer and breast cancer

Abstract The Cancer Genome Atlas and similar projects have now analyzed over 10,000 tumor genomes, providing a catalog of the gene mutations, copy number variants and other genetic alterations that are present cancer. In many cases though, it remains unclear which are the key driver mutations or dependencies in a given cancer and how these influence pathogenesis and response to therapy. Although tumors of similar types and clinical outcomes can have patterns of mutations that are strikingly different, it is becoming apparent that these mutations recurrently hijack the same hallmark molecular pathways and networks. For this reason, cancer research and treatment is increasingly dependent on knowledge of biological networks of multiple types, including physical interactions among proteins and both synthetic-lethal and epistatic interactions among genes. Founded in 2015 by labs at UC San Diego and UCSF, the mission of the Cancer Cell Map Initiative (CCMI) is to enable a new era of cancer discovery and treatment based on the complete elucidation of the molecular networks underlying cancer. We believe that this information will be critical for developing computational models of cancer cells that will enable both basic research and clinical decision-making. Our initial research efforts are focused on head and neck cancer and breast cancer but we believe that our approach is widely applicable. Here, I will discuss results from three of the CCMI's main research efforts. One major focus is using affinity purification followed by mass spectrometry (AP/MS) to map protein-protein interactions of frequently mutated genes in multiple, relevant cell culture models. These efforts for approximately 40 genes in both head and neck cancer and breast cancer have largely been completed leading to a range of validation and follow-up studies. To complement these context-specific structural maps, a second main research effort is to generate functional maps by measuring genetic interactions using CRISPR/Cas9 in many of the same cell culture models. Many of these screens have now been completed with data analysis ongoing. Along with these two experimental efforts, a third focus has been to use bioinformatics approaches to assemble a hierarchical map of biological subsystems in cancer, to identify which systems in which cohorts are mutated more often than expected by chance and to explore these potentially driver systems in more detail through a range of experimental studies. To share these and other network maps, the CCMI's bioinformatics core has created the network data exchange (NDEx, ndexbio.org), an open-source framework where scientists and organizations can share, store, manipulate and publish biological network knowledge. Citation Format: Jason F. Kreisberg, Fan Zheng, Samson Fong, Brent M. Kuenzi, Kyle Ford, Danielle Swaney, Minkyu Kim, Jisoo Park, Zhiyong Wang, Dexter Pratt, Natalia Jura, Silvio Gutkind, Prashant Mali, Nevan Krogan, Trey Ideker. Using physical, genetic and integrated cancer cell maps to investigate head and neck cancer and breast cancer [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 223.