Oncoprotein-specific molecular interaction maps (SigMaps) for cancer network analyses

Abstract

Tumor-specific elucidation of physical and functional oncoprotein interactions could improve tumorigenic mechanism characterization and therapeutic response prediction. Current interaction models and pathways, however, lack context specificity and are not oncoprotein specific. We introduce SigMaps as context-specific networks, comprising modulators, effectors and cognate binding-partners of a specific oncoprotein. SigMaps are reconstructed de novo by integrating diverse evidence sources—including protein structure, gene expression and mutational profiles—via the OncoSig machine learning framework. We first generated a KRAS-specific SigMap for lung adenocarcinoma, which recapitulated published KRAS biology, identified novel synthetic lethal proteins that were experimentally validated in three-dimensional spheroid models and established uncharacterized crosstalk with RAB/RHO. To show that OncoSig is generalizable, we first inferred SigMaps for the ten most mutated human oncoproteins and then for the full repertoire of 715 proteins in the COSMIC Cancer Gene Census. Taken together, these SigMaps show that the cell’s regulatory and signaling architecture is highly tissue specific.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Fig. 1: Protein-specific molecular interaction SigMap and the OncoSigRF algorithm.
Fig. 2: The OncoSigRF LUAD-specific KRAS SigMap.
Fig. 3: Experimental validation of the OncoSigRF LUAD-specific KRAS SigMap.
Fig. 4: KRAS SigMap tumor context specificity.
Fig. 5: OncoSigRF LUAD-specific SigMap analysis of hyper-mutated oncoproteins.
Fig. 6: OncoSig: generalization of OncoSigRF to Cancer Gene Census proteins.

Data availability

All data are provided in Supplementary Files. Source data are provided with this paper.

Code availability

R scripts and code for Naive Bayes and Random Forest classifiers and input data files to reproduce the results described are freely available at https://github.com/califano-lab/OncoSig.

References

  1. 1.

    Prahallad, A. et al. Unresponsiveness of colon cancer to BRAF(V600E) inhibition through feedback activation of EGFR. Nature 483, 100–103 (2012).

    PubMed  CAS  Google Scholar 

  2. 2.

    Bild, A. H. et al. Oncogenic pathway signatures in human cancers as a guide to targeted therapies. Nature 439, 353–357 (2006).

    PubMed  CAS  Google Scholar 

  3. 3.

    Krogan, N. J., Lippman, S., Agard, D. A., Ashworth, A. & Ideker, T. The Cancer Cell Map Initiative: defining the hallmark networks of cancer. Mol. Cell 58, 690–698 (2015).

    PubMed  PubMed Central  CAS  Google Scholar 

  4. 4.

    Greene, C. S. et al. Understanding multicellular function and disease with human tissue-specific networks. Nat. Genet. 47, 569–576 (2015).

    PubMed  PubMed Central  CAS  Google Scholar 

  5. 5.

    Cancer Genome Atlas Research Networket al. The Cancer Genome Atlas Pan-Cancer analysis project. Nat. Genet. 45, 1113–1120 (2013).

    PubMed Central  Google Scholar 

  6. 6.

    Zhang, Q. C. et al. Structure-based prediction of protein–protein interactions on a genome-wide scale. Nature 490, 556–560 (2012).

    PubMed  PubMed Central  CAS  Google Scholar 

  7. 7.

    Garzon, J. I. et al. A computational interactome and functional annotation for the human proteome. eLife 5, e18715 (2016).

    PubMed  PubMed Central  Google Scholar 

  8. 8.

    Ashburner, M. et al. Gene Ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat. Genet. 25, 25–29 (2000).

    PubMed  PubMed Central  CAS  Google Scholar 

  9. 9.

    Margolin, A. A. et al. ARACNE: an algorithm for the reconstruction of gene regulatory networks in a mammalian cellular context. BMC Bioinformatics 7 Suppl 1, S7 (2006).

    PubMed  PubMed Central  Google Scholar 

  10. 10.

    Alvarez, M. J. et al. Functional characterization of somatic mutations in cancer using network-based inference of protein activity. Nat. Genet. 48, 838–847 (2016).

    PubMed  PubMed Central  CAS  Google Scholar 

  11. 11.

    Wang, K. et al. Genome-wide identification of post-translational modulators of transcription factor activity in human B cells. Nat. Biotechnol. 27, 829–839 (2009).

    PubMed  PubMed Central  CAS  Google Scholar 

  12. 12.

    Giorgi, F. M. et al. Inferring protein modulation from gene expression data using conditional mutual information. PLoS ONE 9, e109569 (2014).

    PubMed  PubMed Central  Google Scholar 

  13. 13.

    Network, C. G. A. R. Comprehensive molecular profiling of lung adenocarcinoma. Nature 511, 543–550 (2014).

    Google Scholar 

  14. 14.

    Jansen, R. et al. A Bayesian networks approach for predicting protein–protein interactions from genomic data. Science 302, 449–453 (2003).

    PubMed  CAS  Google Scholar 

  15. 15.

    Breiman, L. Random forests. Machine Learning 45, 5–32 (2001).

    Google Scholar 

  16. 16.

    Liberzon, A. et al. The molecular signatures database (MSigDB) hallmark gene set collection. Cell Syst. 1, 417–425 (2015).

    PubMed  PubMed Central  CAS  Google Scholar 

  17. 17.

    Kanehisa, M., Sato, Y., Kawashima, M., Furumichi, M. & Tanabe, M. KEGG as a reference resource for gene and protein annotation. Nucleic Acids Res. 44, D457–D462 (2016).

    CAS  Google Scholar 

  18. 18.

    Subramanian, A. et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl Acad. Sci. USA 102, 15545–15550 (2005).

    PubMed  PubMed Central  CAS  Google Scholar 

  19. 19.

    Huttlin, E. L. et al. The BioPlex network: a systematic exploration of the human interactome. Cell 162, 425–440 (2015).

    PubMed  PubMed Central  CAS  Google Scholar 

  20. 20.

    Rolland, T. et al. A proteome-scale map of the human interactome network. Cell 159, 1212–1226 (2014).

    PubMed  PubMed Central  CAS  Google Scholar 

  21. 21.

    Chatr-Aryamontri, A. et al. The BioGRID interaction database: 2017 update. Nucleic Acids Res. 45, D369–D379 (2017).

    PubMed  CAS  Google Scholar 

  22. 22.

    Corsello, S. M. et al. The Drug Repurposing Hub: a next-generation drug library and information resource. Nat. Med. 23, 405–408 (2017).

    PubMed  PubMed Central  CAS  Google Scholar 

  23. 23.

    Franceschini, A. et al. STRING v9.1: protein–protein interaction networks, with increased coverage and integration. Nucleic Acids Res. 41, D808–D815 (2013).

    PubMed  CAS  Google Scholar 

  24. 24.

    Lee, I., Blom, U. M., Wang, P. I., Shim, J. E. & Marcotte, E. M. Prioritizing candidate disease genes by network-based boosting of genome-wide association data. Genome Res. 21, 1109–1121 (2011).

    PubMed  PubMed Central  CAS  Google Scholar 

  25. 25.

    Barbie, D. A. et al. Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1. Nature 462, 108–112 (2009).

    PubMed  PubMed Central  CAS  Google Scholar 

  26. 26.

    Kim, J. et al. XPO1-dependent nuclear export is a druggable vulnerability in KRAS-mutant lung cancer. Nature 538, 114–117 (2016).

    PubMed  PubMed Central  CAS  Google Scholar 

  27. 27.

    Corcoran, R. B. et al. Synthetic lethal interaction of combined BCL-XL and MEK inhibition promotes tumor regressions in KRAS mutant cancer models. Cancer Cell 23, 121–128 (2013).

    PubMed  CAS  Google Scholar 

  28. 28.

    Hayes, T. K. et al. Long-term ERK inhibition in KRAS-mutant pancreatic cancer is associated with MYC degradation and senescence-like growth suppression. Cancer Cell 29, 75–89 (2016).

    PubMed  CAS  Google Scholar 

  29. 29.

    Shaw, A. T. et al. Selective killing of K-ras mutant cancer cells by small molecule inducers of oxidative stress. Proc. Natl Acad. Sci. USA 108, 8773–8778 (2011).

    PubMed  CAS  Google Scholar 

  30. 30.

    Liu, Z., Xiao, T., Peng, X., Li, G. & Hu, F. APPLs: more than just adiponectin receptor binding proteins. Cell. Signal. 32, 76–84 (2017).

    PubMed  Google Scholar 

  31. 31.

    Tzeng, H. T. & Wang, Y. C. Rab-mediated vesicle trafficking in cancer. J. Biomed. Sci. 23, 70 (2016).

    PubMed  PubMed Central  Google Scholar 

  32. 32.

    Thomas, J. D. et al. Rab1A is an mTORC1 activator and a colorectal oncogene. Cancer Cell 26, 754–769 (2014).

    PubMed  PubMed Central  CAS  Google Scholar 

  33. 33.

    Gabernet-Castello, C., O’Reilly, A. J., Dacks, J. B. & Field, M. C. Evolution of Tre-2/Bub2/Cdc16 (TBC) Rab GTPase-activating proteins. Mol. Biol. Cell 24, 1574–1583 (2013).

    PubMed  PubMed Central  CAS  Google Scholar 

  34. 34.

    Lu, W. et al. Downregulation of ARHGDIA contributes to human glioma progression through activation of Rho GTPase signaling pathway. Tumour Biol. 37, 15783–15793 (2016).

    PubMed Central  CAS  Google Scholar 

  35. 35.

    Hornstein, I., Alcover, A. & Katzav, S. Vav proteins, masters of the world of cytoskeleton organization. Cell. Signal. 16, 1–11 (2004).

    PubMed  CAS  Google Scholar 

  36. 36.

    Oliver, A. W. et al. The HPV16 E6 binding protein Tip-1 interacts with ARHGEF16, which activates Cdc42. Br. J. Cancer 104, 324–331 (2011).

    PubMed  CAS  Google Scholar 

  37. 37.

    Boulter, E., Estrach, S., Garcia-Mata, R. & Feral, C. C. Off the beaten paths: alternative and crosstalk regulation of Rho GTPases. FASEB J. 26, 469–479 (2012).

    PubMed  CAS  Google Scholar 

  38. 38.

    Cox, A. D. & Der, C. J. Ras history: the saga continues. Small GTPases 1, 2–27 (2010).

    PubMed  PubMed Central  Google Scholar 

  39. 39.

    Prior, I. A. & Hancock, J. F. Ras trafficking, localization and compartmentalized signalling. Semin. Cell Dev. Biol. 23, 145–153 (2012).

    PubMed  CAS  Google Scholar 

  40. 40.

    Bhuin, T. & Roy, J. K. Rab proteins: the key regulators of intracellular vesicle transport. Exp. Cell Res. 328, 1–19 (2014).

    PubMed  CAS  Google Scholar 

  41. 41.

    Fukuda, M. TBC proteins: GAPs for mammalian small GTPase Rab? Biosci. Rep. 31, 159–168 (2011).

    PubMed  CAS  Google Scholar 

  42. 42.

    Hwang, J. & Pallas, D. C. STRIPAK complexes: structure, biological function, and involvement in human diseases. Int. J. Biochem. Cell Biol. 47, 118–148 (2014).

    PubMed  CAS  Google Scholar 

  43. 43.

    Skrzypski, M. et al. Three-gene expression signature predicts survival in early-stage squamous cell carcinoma of the lung. Clin. Cancer Res. 14, 4794–4799 (2008).

    PubMed  CAS  Google Scholar 

  44. 44.

    Li, N. & Li, S. RASAL2 promotes lung cancer metastasis through epithelial–mesenchymal transition. Biochem. Biophys. Res. Commun. 455, 358–362 (2014).

    PubMed  CAS  Google Scholar 

  45. 45.

    Yu, F. et al. IFITM1 promotes the metastasis of human colorectal cancer via CAV-1. Cancer Lett. 368, 135–143 (2015).

    PubMed  CAS  Google Scholar 

  46. 46.

    Weinberg, F. D. & Ramnath, N. Targeting IL22: a potential therapeutic approach for Kras mutant lung cancer? Transl. Lung Cancer Res. 7, S243–S247 (2018).

    PubMed  PubMed Central  Google Scholar 

  47. 47.

    Guillon, A. et al. Interleukin-22 receptor is overexpressed in nonsmall cell lung cancer and portends a poor prognosis. Eur. Respir. J. 47, 1277–1280 (2016).

    PubMed  CAS  Google Scholar 

  48. 48.

    Janne, P. A. et al. Selumetinib plus docetaxel for KRAS-mutant advanced non-small-cell lung cancer: a randomised, multicentre, placebo-controlled, phase 2 study. Lancet Oncol. 14, 38–47 (2013).

    PubMed  Google Scholar 

  49. 49.

    Migliardi, G. et al. Inhibition of MEK and PI3K/mTOR suppresses tumor growth but does not cause tumor regression in patient-derived xenografts of RAS-mutant colorectal carcinomas. Clin. Cancer Res. 18, 2515–2525 (2012).

    PubMed  CAS  Google Scholar 

  50. 50.

    Adjei, A. A. et al. Phase I pharmacokinetic and pharmacodynamic study of the oral, small-molecule mitogen-activated protein kinase kinase 1/2 inhibitor AZD6244 (ARRY-142886) in patients with advanced cancers. J. Clin. Oncol. 26, 2139–2146 (2008).

    PubMed  PubMed Central  CAS  Google Scholar 

  51. 51.

    Sustic, T., Bosdriesz, E., van Wageningen, S., Wessels, L. F. A. & Bernards, R. RUNX2/CBFB modulates the response to MEK inhibitors through activation of receptor tyrosine kinases in KRAS-mutant colorectal cancer. Transl. Oncol. 13, 201–211 (2019).

    PubMed  PubMed Central  Google Scholar 

  52. 52.

    Astsaturov, I. et al. Synthetic lethal screen of an EGFR-centered network to improve targeted therapies. Sci. Signal. 3, ra67 (2010).

    PubMed  PubMed Central  Google Scholar 

  53. 53.

    Cancer Genome Atlas Research Network. Comprehensive molecular profiling of lung adenocarcinoma. Nature 511, 543–550 (2014).

    Google Scholar 

  54. 54.

    Tate, J. G. et al. COSMIC: the catalogue of somatic mutations in cancer. Nucleic Acids Res. 47, D941–D947 (2019).

    CAS  Google Scholar 

  55. 55.

    Narlikar, G. J., Sundaramoorthy, R. & Owen-Hughes, T. Mechanisms and functions of ATP-dependent chromatin-remodeling enzymes. Cell 154, 490–503 (2013).

    PubMed  PubMed Central  CAS  Google Scholar 

  56. 56.

    Morozumi, Y. et al. Atad2 is a generalist facilitator of chromatin dynamics in embryonic stem cells. J. Mol. Cell Biol. 8, 349–362 (2016).

    PubMed  PubMed Central  CAS  Google Scholar 

  57. 57.

    Sharma, V. M., Li, B. & Reese, J. C. SWI/SNF-dependent chromatin remodeling of RNR3 requires TAF(II)s and the general transcription machinery. Genes Dev. 17, 502–515 (2003).

    PubMed  PubMed Central  CAS  Google Scholar 

  58. 58.

    Mi, H., Muruganujan, A., Ebert, D., Huang, X. & Thomas, P. D. PANTHER version 14: more genomes, a new PANTHER GO-slim and improvements in enrichment analysis tools. Nucleic Acids Res. 47, D419–D426 (2019).

    PubMed  CAS  Google Scholar 

  59. 59.

    Gong, F. & Miller, K. M. Double duty: ZMYND8 in the DNA damage response and cancer. Cell Cycle 17, 414–420 (2018).

    PubMed  PubMed Central  CAS  Google Scholar 

  60. 60.

    Sridhara, S. C. et al. Transcription dynamics prevent RNA-mediated genomic instability through SRPK2-dependent DDX23 phosphorylation. Cell Rep. 18, 334–343 (2017).

    PubMed  CAS  Google Scholar 

  61. 61.

    Allemand, E. et al. A broad set of chromatin factors influences splicing. PLoS Genet. 12, e1006318 (2016).

    PubMed  PubMed Central  Google Scholar 

  62. 62.

    Blume-Jensen, P. & Hunter, T. Oncogenic kinase signalling. Nature 411, 355–365 (2001).

    PubMed  CAS  Google Scholar 

  63. 63.

    Lemmon, M. A. & Schlessinger, J. Cell signaling by receptor tyrosine kinases. Cell 141, 1117–1134 (2010).

    PubMed  PubMed Central  CAS  Google Scholar 

  64. 64.

    Organ, S. L. & Tsao, M. S. An overview of the c-MET signaling pathway. Ther. Adv. Med. Oncol. 3, S7–S19 (2011).

    PubMed  PubMed Central  CAS  Google Scholar 

  65. 65.

    Meissl, K., Macho-Maschler, S., Muller, M. & Strobl, B. The good and the bad faces of STAT1 in solid tumours. Cytokine 89, 12–20 (2017).

    PubMed  CAS  Google Scholar 

  66. 66.

    Zhang, Y. & Liu, Z. STAT1 in cancer: friend or foe? Discov. Med. 24, 19–29 (2017).

    PubMed  Google Scholar 

  67. 67.

    Balbin, O. A. et al. Reconstructing targetable pathways in lung cancer by integrating diverse omics data. Nat. Commun. 4, 2617 (2013).

    PubMed  PubMed Central  Google Scholar 

  68. 68.

    Downward, J. RAS synthetic lethal screens revisited: still seeking the elusive prize? Clin. Cancer Res. 21, 1802–1809 (2015).

    PubMed  PubMed Central  CAS  Google Scholar 

  69. 69.

    Luo, J. et al. A genome-wide RNAi screen identifies multiple synthetic lethal interactions with the Ras oncogene. Cell 137, 835–848 (2009).

    PubMed  PubMed Central  CAS  Google Scholar 

  70. 70.

    Wang, T. et al. Gene essentiality profiling reveals gene networks and synthetic lethal interactions with oncogenic Ras. Cell 168, 890–903 (2017).

    PubMed  PubMed Central  CAS  Google Scholar 

  71. 71.

    McDonald, E. R. III et al. Project DRIVE: a compendium of cancer dependencies and synthetic lethal relationships uncovered by large-scale, deep RNAi screening. Cell 170, 577–592 (2017).

    CAS  Google Scholar 

  72. 72.

    Aguirre, A. J. & Hahn, W. C. Synthetic lethal vulnerabilities in KRAS-mutant cancers. Cold Spring Harb. Perspect. Med. 8, a031518 (2018).

    PubMed  PubMed Central  Google Scholar 

  73. 73.

    Woo, J. H. et al. Elucidating compound mechanism of action by network perturbation analysis. Cell 162, 441–451 (2015).

    PubMed  PubMed Central  CAS  Google Scholar 

  74. 74.

    Duan, Q. et al. LINCS Canvas Browser: interactive web app to query, browse and interrogate LINCS L1000 gene expression signatures. Nucleic Acids Res. 42, W449–W460 (2014).

    PubMed  PubMed Central  CAS  Google Scholar 

  75. 75.

    Kramer, A., Green, J., Pollard, J. Jr. & Tugendreich, S. Causal analysis approaches in Ingenuity Pathway Analysis. Bioinformatics 30, 523–530 (2014).

    PubMed  Google Scholar 

  76. 76.

    Li, W. & Godzik, A. Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences. Bioinformatics 22, 1658–1659 (2006).

    PubMed  CAS  Google Scholar 

  77. 77.

    Arlot, S. & Celisse, A. A survey of cross-validation procedures for model selection. Statist. Surv. 4, 40–79 (2010).

    Google Scholar 

  78. 78.

    Torres, J. Z., Miller, J. J. & Jackson, P. K. High-throughput generation of tagged stable cell lines for proteomic analysis. Proteomics 9, 2888–2891 (2009).

    PubMed  PubMed Central  CAS  Google Scholar 

  79. 79.

    Shevchenko, A., Tomas, H., Havlis, J., Olsen, J. V. & Mann, M. In-gel digestion for mass spectrometric characterization of proteins and proteomes. Nat. Protoc. 1, 2856–2860 (2006).

    PubMed  CAS  Google Scholar 

  80. 80.

    Zheng, Y. et al. A rare population of CD24(+)ITGB4(+)Notch(hi) cells drives tumor propagation in NSCLC and requires Notch3 for self-renewal. Cancer Cell 24, 59–74 (2013).

    PubMed  PubMed Central  CAS  Google Scholar 

  81. 81.

    Dai, Z. et al. edgeR: a versatile tool for the analysis of shRNA-seq and CRISPR–Cas9 genetic screens. F1000Res 3, 95 (2014).

    PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

This work was supported by the National Cancer Institute (NCI) Outstanding Investigator Award R35CA197745 to A.C.; the NCI Cancer Target Discovery and Development Program U01CA168426 to A.C.; the NCI Research Centers for Cancer Systems Biology Consortium U54CA209997 to A.C. and B.H.; National Institute of General Medical Sciences grant R01GM30518 to B.H.; NCI grant R01CA129562 to E.A.S.C.; Innovative Research Grant from Stand Up to Cancer to E.A.S.C.; National Institutes of Health High-End Instrumentation Program grant S10OD012351 to A.C.; and NIH Shared Instrumentation Program grant S10OD021764 to A.C. J.B. was supported, in part, by the Ruth L. Kirschstein National Research Service Award Institutional Research Training Grant T32GM082797. D.R.S. was supported by the Ruth L. Kirschstein National Research Service Award Institutional Research Training Grant T32CA09302. Relevant ethical compliance was ensured by the Institutional Review Board of the Human Research Protection Program at the University of California, San Francisco.

Author information

Affiliations

Authors

Contributions

J.B., D.M., A.T.G., A.L. and F.M.G. performed computational analysis. J.B. performed machine learning. D.S. and P.K.J. performed, respectively, the knockdown and AP–MS experiments. E.O.P, B.W.C. and S.T. created oncosig.org. J.B., D.M. and S.J.J. compiled the codebook. J.B., D.S., E.A.S.C., A.C. and B.H. analyzed knockdown experiments. S.D.M., B.H., A.C. and E.A.S.C. designed research. D.M., B.H. and A.C. designed computational and experimental work and analyzed data. D.M., D.S., J.B., A.C. and B.H. assembled the data and wrote the paper.

Corresponding authors

Correspondence to E. Alejandro Sweet-Cordero or Barry Honig or Andrea Califano.

Ethics declarations

Competing interests

A.C. is the founder and an equity holder in DarwinHealth, a company that has licensed some of the algorithms used in this study from Columbia University. Columbia University is also an equity holder in DarwinHealth.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 The top 40 predictions from the OncoSigNB algorithm for the KRAS LUAD SigMap were chosen for validation.

a, Performance of OncoSigNB at recovering the Ingenuity-derived positive gold standard set (PGSS) as a function of LRPost. At an LRPost = 53 (probability = 0.50) (vertical gray line), the OncoSigNB LUAD-specific KRAS SigMap contains 10% of the PGSS (horizontal gray line). The vertical red line corresponds to LRPost = 240, the cutoff used to obtain candidates for experimental validation. b, ROC curve analysis, evaluated as the recovery of the Ingenuity-derived PGSS (FPR ≤ 0.05), for 1) OncoSigNB (green curve, N = 1028)), 2) Pearson’s correlation between mRNA expression of KRAS and mRNA expression of other proteins in LUAD (blue curve), and 3) random performance (black curve). Recovery using 2-fold cross-validation (green) is essentially indistinguishable from recovery using 100-fold Monte-Carlo Cross-validation (not shown). 393 OncoSigNB LUAD-specific KRAS SigMap predictions are made for LRPost ≥ 53, which corresponds to probability ≥ 0.50 and FPR ≤ 0.019 (purple dot). 40 OncoSigNB LUAD-specific KRAS SigMap predictions are made for LRPost ≥ 240, which corresponds to probability ≥ 0.82 and FPR ≤ 0.0018 (yellow dot). The top 40 predictions are listed by gene name in (c). c, Orange and blue boxes contain, respectively, known upstream regulators and downstream effectors that are successfully recovered by OncoSigNB. Italicized text indicates proteins known to interact with KRAS via a physical protein-protein interaction. The box titled “validated predictions” shows the novel OncoSigNB predictions tested with the RNAi negative screen; those that were experimentally found to affect cell growth in a KRAS-dependent context are highlighted in bold text. Source data

Extended Data Fig. 2 The OncoSigRF and OncoSigNB algorithms produce highly similar KRAS LUAD SigMaps.

a, Comparison of ROC curves (FPR ≤ 0.05) for LUAD-specific KRAS SigMaps predicted by OncoSigNB (green and blue curves) and OncoSigRF (orange and red curve) trained on the Ingenuity PGSS and the MSigDB PGSS, respectively. b, Gene Set Enrichment Analysis (GSEA) of the top 100 OncoSigNB LUAD-specific KRAS SigMap predictions at the top of the OncoSigRF LUAD-specific KRAS SigMap predictions. Ranking is based on OncoSigRF score (SRF). Both the OncoSigNB predictions tested in the knockdown experiments (red lines) and the remaining top 100 OncoSigNB predictions (blue lines) are highly enriched at the top of the OncoSigRF predictions (p = 5.6 ×10−8 and p = 1.7 ×10-19, respectively). Source data

Extended Data Fig. 3 The log2FC of shRNA abundance is plotted against the novel proteins tested in the KRAS negative selection screen.

The 3–5 points plotted for a given protein are shRNAs that target the mRNA for that protein (N = 100). The X-axis is sorted by mean log2FC for all shRNAs targeting each gene. Colors change from red to green with mean log2FC.

Extended Data Fig. 4 OncoSigRF predictions are highly enriched in oncogenic KRASMut dependencies.

a, GSEA of KRASMut synthetic lethal partners25 (blue lines, N = 216) and the top 500 OncoSigRF LUAD-specific KRAS SigMap predictions obtained by training on a modified PGSS for which the intersection with the synthetic lethal set was removed. Inset is the GSEA using all OncoSigRF predictions obtained in this way, where the ranking is OncoSigRF score. Enrichment analysis was performed with the aREA (analytic Rank-based Enrichment Analysis) algorithm10. b, Enrichment of the protein resistance-signature to ERK inhibitor SCH77298428 (blue lines, N = 24) within OncoSigRF LUAD-specific KRAS SigMap predictions10. c, Enrichment of proteins involved in response to Reactive Oxygen Species (GO:0000302)8,29 (blue lines, N = 276) within OncoSigRF LUAD-specific KRAS SigMap predictions10.

Extended Data Fig. 5 OncoSigRF KRAS SigMaps exhibit tissue context specificity.

a, ROC curves for the OncoSigRF KRAS SigMaps in LUAD (red), LUSC (gray) COAD (brown), and PAAD (orange) for FPR ≤ 0.05. Performance is evaluated as the recovery of established KRAS pathway proteins. b, Gene set enrichment analysis (GSEA) of KRASMut synthetic lethal partners, as determined by Corcoran et al27 (N = 48, blue lines). To avoid training and testing on the same proteins, OncoSigRF predictions for COAD-specific KRAS SigMap proteins were obtained by training on a modified PGSS from which any established KRASMut synthetic lethal protein had been previously removed. Enrichment analysis was performed with the aREA (analytic Rank-based Enrichment Analysis) algorithm10. c, Scatterplot of OncoSigRF scores for KRAS SigMap proteins in PAAD-vs-LUAD (N = 19,789). Each dot represents the scores for one protein. Darker colored points have high scores (SRF ≥ 0.5) in at least one context, and lighter colored points score poorly in both contexts (SRF ≤ 0.5). R2PAAD/LUAD = 0.037. d, OncoSigRF COAD-specific KRAS SigMap in the form depicted conceptually in Fig. 1a. To prevent visual cluttering, only the top 33 OncoSigRF predictions (FPR ≤ 0.01) that are also VIPER-inferred KRAS interactors (p ≤ 0.01), PrePPI-predicted KRAS physical interactors, or both, are depicted. Bold and regular text node labels represent established and novel predictions, respectively; orange and blue node colors represent upstream regulators and downstream effectors, respectively; red, blue, and black node borders represent predictions that are druggable (Drug Repurposing Hub22), KRASMut synthetic lethal from the literature and validated here (see text), and both, respectively; orange and blue solid lines and gray nodes represent PrePPI-predicted physical interactors of KRAS. Source data

Extended Data Fig. 6 OncoSigRF SigMaps for hypermutated oncoproteins are retrospectively validated.

a, Pairwise overlap of established pathway proteins (left) and the OncoSigRF LUAD-specific SigMaps (FPR ≤ 0.01, right) for the ten hyper-mutated oncoproteins (names of columns and rows). Percent overlap is color-coded according to the scale at top. b, SigMap predictions are highly enriched in 600 EGFR-centric network proteins52 (p = 2.3 ×10−43). Enrichment analysis was performed with the aREA (analytic Rank-based Enrichment Analysis) algorithm10. c, Box plots of the OncoSigRF LUAD-specific EGFR SigMap scores for two subsets of the curated EGFR pathway proteins from Astsaturov et al:52 those identified as EGFR synthetic lethal partners (red, N = 58) and those not identified as synthetic lethal (grey, N = 542). The p-value (2 ×10−4) was calculated using Welch’s two sample t-test. Source data

Supplementary information

Source data

Source Data Fig. 1c

Statistical Source Data

Source Data Fig. 2a

Statistical Source Data

Source Data Fig. 2b

Statistical Source Data

Source Data Fig. 3d

Statistical Source Data

Source Data Fig. 4a

Statistical Source Data

Source Data Fig. 4b

Statistical Source Data

Source Data Fig. 5a

Statistical Source Data

Source Data Fig. 5b

Statistical Source Data

Source Data Fig. 6a

Statistical Source Data

Source Data Fig. 3d

Layered vector image Powerpoint

Source Data Fig. 6bcd

Layered vector image Powerpoint

Source Data Extended Data Fig. 1a

Statistical Source Data

Source Data Extended Data Fig. 1b

Statistical Source Data

Source Data Extended Data Fig. 2a

Statistical Source Data

Source Data Extended Data Fig. 5a

Statistical Source Data

Source Data Extended Data Fig. 5c

Statistical Source Data

Source Data Extended Data Fig. 6a

Statistical Source Data

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Broyde, J., Simpson, D.R., Murray, D. et al. Oncoprotein-specific molecular interaction maps (SigMaps) for cancer network analyses. Nat Biotechnol (2020). https://doi.org/10.1038/s41587-020-0652-7

Download citation

Search

Quick links

Sign up for the Nature Briefing newsletter for a daily update on COVID-19 science.
Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing