TL;DR
This paper introduces two Cytoscape-based tools, cyTRON and cyTRON/JS, for visualizing and manipulating cancer evolution models derived from genomic data, enhancing cancer research capabilities.
Contribution
The paper presents two novel applications, one in Java and one in JavaScript/R, for cancer evolution model inference and visualization, integrating with public databases.
Findings
Two new Cytoscape applications for cancer evolution modeling
Tools are publicly available and easy to integrate with genomic data
Enhanced visualization and interaction with cancer progression models
Abstract
The increasing availability of sequencing data of cancer samples is fueling the development of algorithmic strategies to investigate tumor heterogeneity and infer reliable models of cancer evolution. We here build up on previous works on cancer progression inference from genomic alteration data, to deliver two distinct Cytoscape-based applications, which allow to produce, visualize and manipulate cancer evolution models, also by interacting with public genomic and proteomics databases. In particular, we here introduce cyTRON, a stand-alone Cytoscape app, and cyTRON/JS, a web application which employs the functionalities of Cytoscape/JS. cyTRON was developed in Java; the code is available at https://github.com/BIMIB-DISCo/cyTRON and on the Cytoscape App Store http://apps.cytoscape.org/apps/cytron. cyTRON/JS was developed in JavaScript and R; the source code of the tool is available at…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
cyTRON and cyTRON/JS: two Cytoscape-based applications for the inference of cancer evolution models
Lucrezia Patruno111Equal contributors.
Department of Informatics, Systems and Communication, University of Milano-Bicocca, Milan, Italy
Edoardo Galimberti*∗*
Department of Computer Science, University of Turin, Turin, Italy
ISI Foundation, Turin, Italy
Daniele Ramazzotti
Department of Pathology, Stanford University, Stanford, CA, United States
Giulio Caravagna
The Institute of Cancer Research - ICR, London, UK
Luca De Sano
Department of Informatics, Systems and Communication, University of Milano-Bicocca, Milan, Italy
Marco Antoniotti
Department of Informatics, Systems and Communication, University of Milano-Bicocca, Milan, Italy
Milan Center for Neuroscience, Milan, Italy
Alex Graudenzi
Department of Informatics, Systems and Communication, University of Milano-Bicocca, Milan, Italy
Institute of Molecular Bioimaging and Physiology - IBFM of the National Research Council - CNR, Segrate, Milan, Italy
Abstract
The increasing availability of sequencing data of cancer samples is fueling the development of algorithmic strategies to investigate tumor heterogeneity and infer reliable models of cancer evolution. We here build up on previous works on cancer progression inference from genomic alteration data, to deliver two distinct Cytoscape-based applications, which allow to produce, visualize and manipulate cancer evolution models, also by interacting with public genomic and proteomics databases. In particular, we here introduce cyTRON, a stand-alone Cytoscape app, and cyTRON/JS, a web application which employs the functionalities of Cytoscape/JS.
cyTRON was developed in Java; the code is available at
https://github.com/BIMIB-DISCo/cyTRON and on the Cytoscape App Store http://apps.cytoscape.org/apps/cytron. cyTRON/JS was developed in JavaScript and R; the source code of the tool is available at https://github.com/BIMIB-DISCo/cyTRON-js and the tool is accessible from https://bimib.disco.unimib.it/cytronjs/welcome.
1 Scientific Background
Cancer is a complex disease, whose development is caused by the accumulation of alterations in the genome. Some alterations may confer a selective advantage to cancer cells, and this may result in the expansion of cancer clones. In order to understand how cancer evolves, it is of great importance to understand how such driver alterations accumulate over time [Nowell, 1976, Burrell et al., 2013]. This goal can be pursued by reconstructing cancer evolution models, which are graphs that encode the evolutionary history of drivers and their temporal relationships. The reconstruction of such models is a complex task mainly because of two reasons: first, much of the data publicly available from initiatives such as TCGA [https://portal.gdc.cancer.gov/] comes from cross-sectional samples, and hence they lack of temporal information. The second main reason can be found in the heterogeneity of tumors [Caravagna et al., 2016, Ramazzotti et al., 2018].
2 Materials and Methods
In order to learn meaningful evolution models, we developed a pipeline, PICNIC [Caravagna et al., 2016], which includes the following steps: identification of homogeneous sample subgroups (e.g., tumor subtypes), identification of drivers (i.e., the nodes of the output model), identification of mutually exclusive patterns, ) inference of cancer evolution models via distinct algorithms (e.g., [Loohuis et al., 2014, Ramazzotti et al., 2015, Ramazzotti et al., 2019]). This pipeline was implemented within the widely used TRONCO R suite for TRanslational ONCOlogy [De Sano et al., 2016], [Antoniotti et al., 2015], which was recently employed, for instance, to analyze the largest kidney cancer cohort currently available [Turajlic et al., 2018].
However, TRONCO presents two practical limitations: first, it requires at least some basic programming skills due to its underlying R infrastructure; second, TRONCO is not integrated with publicly available genomic databases, hence providing a non-interactive visualization of the output graphs.
Therefore, to improve the practicality, effectiveness, interactivity and diffusion of our framework, we integrated it within Cytoscape, an user-friendly open-source platform for the visualization and manipulation of complex networks [Shannon et al., 2003]. We here present cyTRON, a stand-alone Cytoscape app, and cyTRON/JS a web application which employs the functionalities of Cytoscape/JS, both of which allow to produce, visualize and manipulate cancer evolution models, also by interacting with public genomic databases. Figure 1B shows an example of the output in cyTRON/JS, which exploits Cytoscape/JS to provide an interactive visualization of the evolution model.
cyTRON and cyTRON/JS were designed for two main purposes:
- •
Providing an interactive and user-friendly visualization of TRONCO models: while TRONCO R-based graph display is static, cyTRON and cyTRON/JS provide interactive views, which allow to directly retrieve information about genes involved in the study, by accessing widely-used public genome databases.
- •
Making TRONCO accessible to users unfamiliar with R programming: cyTRON and cyTRON/JS provide interfaces that enable the usage of TRONCO respectively from Cytoscape and a Web browser, thus removing the need for users to execute any code in order to complete a whole analysis.
The architecture of both tools can be conceptually defined as follows:
- •
The front-end side is composed of an interface that can be used to:
Select input data for the TRONCO analysis [De Sano et al., 2016]: the input files should be either MAF, GISTIC or user-defined Boolean matrices that contain information about the mutations observed in each sample. 2. 2.
Set the parameters for the inference in order to access most TRONCO capabilities. Users can specify which driver mutations to include in the analysis, which algorithm among those implemented in TRONCO to use for the reconstruction and the algorithm’s corresponding parameters. 3. 3.
Visualize cancer evolution models and dynamically interact with the result: for instance, by clicking on the genes of the output graph, it is possible to retrieve the information available on public genomic databases. cyTRON gives access to gene information through databases such as Ensembl222http://www.ensembl.org/index.html and Entrez333https://www.ncbi.nlm.nih.gov/search/, that are accessible through the Cytoscape interface.
In cyTRON/JS the data displayed for each node are retrieved from the Entrez database Gene 444https://www.ncbi.nlm.nih.gov/gene/ using E-Utils, an API provided by the National Center for Biotechnology Information.
- •
The back-end side includes the communication channel with R. For cyTRON, a Java bridge with R is built by means of rJava. Instead, for cyTRON/JS it is based on js-call-r, a Node.js package which collects the data and parameters set by the user, encodes them in JSON and sends them to R. Then, R commands are transparently executed in order to perform any specific step of the analysis by TRONCO.
Figure 1 shows a view of cyTRON workspace (left) and an example of output model by cyTRON/JS (right). In order to choose between the two tools, users should take into consideration the data and the type of analysis they need to carry out. In particular, since cyTRON/JS is a web application, it is readily accessible from any device and computations are carried out on the back-end side. This feature is useful in case a user needs to carry out a computational-expensive analysis. However, cyTRON is more complete with respect to all the functionalities implemented in TRONCO: for example, it implements also the option of testing hypothesis on mutations through the algorithm Capri [Ramazzotti et al., 2015].
3 Conclusion and future work
TRONCO is an R package that implements state-of-the-art algorithms for the inference of cancer evolution models with the ultimate goal of understanding the evolutionary trajectories driving tumor evolution. In such a multidisciplinary domain, where computer scientists actively cooperate with biologists, being capable of visually understanding the data is crucial to both parties. In order to effectively allow the use of TRONCO, we here presented cyTRON and cyTRON/JS, two Cytoscape-based applications which translate many of TRONCO functionalities into Cytoscape. Our effort aims at designing user-friendly and accessible tools to support the user in the task of exploring cancer genomic data.
As the TRONCO functionalities are constantly updated and improved, these two new tools need to be kept up to date. Thus, future work will be focused on integrating TRONCO’s new algorithms for analyzing single cell datasets [Ramazzotti et al., 2019]. In addition to this, cyTRON/JS needs to be extended with the hypothesis testing functionality, in order to enable users to carry out more complex analysis.
Acknowledgments
cyTRON and cyTRON/JS were developed within the Google summer of Code program, in collaboration with the NRNB organization. The authors declare no conflicts of interest.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1[Antoniotti et al., 2015] Antoniotti, M., Caravagna, G., De Sano, L., Graudenzi, A., Mauri, G., Mishra, B., and Ramazzotti, D. (2015). Design of the tronco bioconductor package for translational oncology. Submitted. Available on bio Rxiv. org, ID , 27524.
- 2[Burrell et al., 2013] Burrell, R. A., Mc Granahan, N., Bartek, J., and Swanton, C. (2013). The causes and consequences of genetic heterogeneity in cancer evolution. Nature , 501(7467):338.
- 3[Caravagna et al., 2016] Caravagna, G., Graudenzi, A., Ramazzotti, D., Sanz-Pamplona, R., De Sano, L., Mauri, G., Moreno, V., Antoniotti, M., and Mishra, B. (2016). Algorithmic methods to infer the evolutionary trajectories in cancer progression. Proceedings of the National Academy of Sciences , 113(28):E 4025–E 4034.
- 4[De Sano et al., 2016] De Sano, L., Caravagna, G., Ramazzotti, D., Graudenzi, A., Mauri, G., Mishra, B., and Antoniotti, M. (2016). Tronco: an r package for the inference of cancer progression models from heterogeneous genomic data. Bioinformatics , 32(12):1911–1913.
- 5[Loohuis et al., 2014] Loohuis, L. O., Caravagna, G., Graudenzi, A., Ramazzotti, D., Mauri, G., Antoniotti, M., and Mishra, B. (2014). Inferring tree causal models of cancer progression with probability raising. Plo S one , 9(10):e 108358.
- 6[Nowell, 1976] Nowell, P. C. (1976). The clonal evolution of tumor cell populations. Science , 194(4260):23–28.
- 7[Ramazzotti et al., 2015] Ramazzotti, D., Caravagna, G., Olde Loohuis, L., Graudenzi, A., Korsunsky, I., Mauri, G., Antoniotti, M., and Mishra, B. (2015). Capri: efficient inference of cancer progression models from cross-sectional data. Bioinformatics , 31(18):3016–3026.
- 8[Ramazzotti et al., 2018] Ramazzotti, D., Graudenzi, A., Caravagna, G., and Antoniotti, M. (2018). Modeling cumulative biological phenomena with suppes-bayes causal networks. Evolutionary Bioinformatics , 14:1176934318785167.
