Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2022 Dec 12;22(2):368–373. doi: 10.1021/acs.jproteome.2c00512

CPPA: A Web Tool for Exploring Proteomic and Phosphoproteomic Data in Cancer

Guo-sheng Hu †,‡, Zao-zao Zheng †,‡, Yao-hui He †,‡, Du-chuang Wang †,‡, Wen Liu †,‡,*
PMCID: PMC9904288  PMID: 36507870

Abstract

graphic file with name pr2c00512_0003.jpg

A tremendous amount of proteomic and phosphoproteomic data has been produced over the years with the development of mass spectrometry techniques, providing us with new opportunities to explore and understand the proteome and phosphoproteome as well as the function of proteins and protein phosphorylation sites. However, a lack of powerful tools that we can utilize to explore these valuable data limits our understanding of the proteome and phosphoproteome, particularly in diseases such as cancer. To address these unmet needs, we established CPPA (Cancer Proteome and Phosphoproteome Atlas), a web tool to mine abnormalities of the proteome and phosphoproteome in cancer based on published data sets. All analysis results are presented in CPPA with a flexible web interface to provide key customization utilities, including general analysis, differential expression profiling, statistical analysis of protein phosphorylation sites, correlation analysis, similarity analysis, survival analysis, pathological stage analysis, etc. CPPA greatly facilitates the process of data mining and therapeutic target discovery by providing a comprehensive analysis of proteomic and phosphoproteomic data in normal and tumor tissues with a simple click, which helps to unlock the precious value of mass spectrometry data by bridging the gap between raw data and experimental biologists. CPPA is currently available at https://cppa.site/cppa.

Keywords: proteome, phosphoproteome, cancer, web tool

Introduction

Large-scale omics data resources, such as the International Cancer Genome Consortium (ICGC)1 and the Cancer Genome Atlas (TCGA),2 have greatly helped us understand the characteristics of a wide variety of cancers. Through these data resources, molecular aberrations at DNA, RNA, and epigenetic levels can be systematically identified for screening novel biomarkers or candidate drug targets.3,4 The study of genomics and transcriptomics is the first step toward precision oncology, and proteomics has, in many quarters, been considered the next logical step in expanding our understanding of tumor biology because it provides information that complements genomic and transcriptomic data.5 Clinical Proteomic Tumor Analysis Consortium (CPTAC) is a project that aims to accelerate the understanding of the molecular basis of cancer through the application of large-scale proteome and genome analysis.6 CPTAC has so far produced proteomic and phosphoproteomic data across 15 different types of tumors, covering over 2000 patients. However, the statistical analysis and data visualization provided by the data portal of CPTAC are not convenient enough for researchers with limited bioinformatics skills, thus limiting the potential values of CPTAC data sets. Therefore, it is important to develop user-friendly tools to analyze and visualize the data sets in CPTAC. A key step toward understanding cancer is to analyze the changes of proteins as well as post-translational modifications (PTMs) such as phosphorylation in tumor tissues through proteomic and phosphoproteomic data. The differential expression of proteins between normal and tumor tissues is generally analyzed through statistical methods.7 Many potential biomarkers and therapeutic targets have been identified based on abnormally expressed proteins in a variety of cancers, including esophageal cancer,8 breast cancer,9 colorectal cancer,10 liver cancer,11 and so on. Besides the changes in abundance, proteins generate tremendous diversity, complexity, and heterogeneity through PTMs. There are currently over 300 different types of PTMs, but only a few have been studied in depth,12 including phosphorylation, methylation, acetylation, ubiquitin, glycosylation, etc. How to systematically and comprehensively study these PTMs is one of the main challenges in proteomics research. Protein phosphorylation, principally on serine, threonine, or tyrosine residues, is one of the most important and well-studied post-translational modifications. Protein phosphorylation plays a critical role in both physiological and pathological conditions.13 Therefore, proteomic and phosphoproteomic data may provide precious insights into tumor biology that cannot be deciphered by genomic analysis.

Currently, there are many excellent web tools for statistical analysis and data visualization of cancer genomic and transcriptomic data, including cBioPortal,14 GEPIA,15 UALCAN,16,17 and so on. cBioPortal is focused on tumor samples and provides an interface for cancer genomic data download and visualization. GEPIA provides interactive analysis of transcriptome data from TCGA and GTEx.18,19 UALCAN provides multiform data analysis by using transcriptomic data well as pan-cancer expression analysis based on proteomic data. In addition, HPA provides an atlas analysis of tissue, blood, cells, and organs based on proteomic, transcriptomic, and system biology data resources.20 HPA can also provide protein abundance analysis according to immunohistochemistry, but is limited to tumor tissues. LinkedOmics provides interactive analysis across 10 cancer types based on proteomic and phosphoproteomic data, but the operating logic is complex and must specify one gene as the core in the analysis.21 Taken together, none of the tools can take full advantage of the proteomic and phosphoproteomic data from CPTAC or other published data sets for systematic statistical analysis and visual exhibition of proteins and phosphorylation sites in both normal and cancer samples. To address these unmet needs, we developed CPPA (Cancer Proteome and Phosphoproteome Atlas), a comprehensive web tool to identify aberrant patterns of proteins and phosphorylation sites that are correlated to clinical records in cancer.

Methods and Results

Data Collection

The data sets included in CPPA were collected from published data sets (Table S1), most of which were downloaded from CPTAC, which encompasses mass spectrometry (MS)-based proteomic data, phosphoproteomic data, and clinical records from over 2000 cancer patients. The operation processes of the proteome and phosphoproteome based on mass spectrometry are complex, and the quality of data generated by different institutions or laboratories could be largely different. Considering the current situation of proteomic data and the intention of the website design herein, five criteria were applied for screening the data sets. First, proteomic data should be labeled with isobaric tandem mass tags, such as isobaric tags for relative and absolute quantification reagents (iTRAQ) and tandem mass tag (TMT) reagents, which improve the data quality and reproducibility. Second, phosphoproteomic data must be produced together with the proteomic data using the same batch of samples. Third, tumor tissues must be collected before the patients are treated. Fourth, normal samples must be produced in the same project based on the predesigned prototype of this database. Fifth, the number of samples must be greater than 50. The final data sets included in CPPA contained 13 different types of cancer samples, covering nearly 1500 patients.

Data Preprocessing

Proteomic and phosphoproteomic raw data were downloaded from the CPTAC data portal or PRIDE (https://www.ebi.ac.uk/pride/). All proteomic and phosphoproteomic raw data from different sources have been recomputed through a standard pipeline to minimize the abiologic difference.

MaxQuant software (version 2.0.2.0) was used to analyze MS raw files.22 MS/MS spectra were searched against the reviewed SwissProt human proteome database and a common contaminants database by the Andromeda search engine.23 If the “internal reference”, such as a mixed sample, exists, the reference channel will be set according to the corresponding plex, and the normalization method will be set as “Weighted ratio to reference channel”.24 Carbamidomethylation was applied as fixed and N-terminal acetylation, deamidation at NQ, and methionine oxidation as variable modifications. Phosphorylation of serine, threonine, and tyrosine (S, T, and Y) was set as variable modifications in analysis pipeline of phosphoproteomic raw data. Enzyme specificity was set to “Trypsin/P” or “Trypsin/P + LysC” with a maximum of 2 missed cleavages and a minimum peptide length of 7 amino acids according to corresponding published papers. A false discovery rate (FDR) of 1% was applied at the peptide and protein level. Peptide identification was performed with an allowed initial precursor mass deviation of up to 7 ppm and an allowed fragment mass deviation of 20 ppm. Protein identification required at least 1 “razor + unique peptides”. Data were filtered for common contaminants in proteomic data and phosphoproteomic data, and protein groups only identified by peptides with modifications were excluded from further analysis in proteomic data.

Standard pipeline analysis of proteomic raw data consumed approximately 100 000 h of CPU time, while the phosphoproteomics analysis pipeline took nearly 600 000 h of CPU time to run out.

To mitigate systematic and sample-specific bias in the quantification, the expression ratios were log2-transformed and normalized using the median centering method across proteins and phosphorylation sites.

Finally, we generated a structured table containing the information about the number of samples and identified proteins and phosphorylation sites in each data set (Table S1).

Clinical information was downloaded from the CPTAC data portal or obtained from published papers. Major clinical parameters, including clinical stage, pathological stage, histological grade, age, gender, survival time, and vital status of all cancer patients, were further organized into structured data tables.

Implementations

The web tool was hosted by Nginx in ubuntu 20.04. The back end was implemented using Flask v2.0.2. All processed data described above were imported and stored into SQLite relational database. The front end was built using HTML, CSS, and JavaScript, including the Bootstrap framework (v4) for the styling and jQuery for HTML scripts. Statistical results are generated by Python or R. Analysis results in CPPA were exhibited by tables or plots. Tables are created by Bootstrap Table (https://bootstrap-table.com/) and plots are generated by plotly.js library (https://plot.ly/) or R package ggplot2. Survival analysis is generated by R package survminer (Figure 1).

Figure 1.

Figure 1

Open in a new tab

Schema describing data processing and data display for the CPPA web tool.

Functionalities

Functionalities of CPPA are divided into multiple major modules: General analysis, Differential profile, Protein phosphorylation analysis, Custom expression analysis, Correlation analysis, Similarity analysis, and Survival analysis (Figure 2). The “Help” webpage, provided by CPPA website, describes the full analysis tutorial and detailed user manual (https://cppa.site/cppa/tppa_help).

Figure 2.

Figure 2

Open in a new tab

Examples of CPPA interactive results. (A) Users can query abundance of proteins (upper panel) and distribution of phosphorylation sites (bottom panel) by “General” module. (B) Differential expression analysis is displayed by pie chart (upper panel, left) and volcano plot (upper panel, right) in CPPA. The normalized expression level, fold change, and Q value is shown in a table (bottom panel). (C) Distribution (upper panel) and abundance of phosphorylation sites (bottom panel) in proteins are displayed through lollipop plot and heat map, respectively. (D) Customized expression analysis (left panel), pathological stage analysis (right panel, upper), and comparison analysis (right panel, bottom) are exhibited through box plot, violin plot, and heat map, respectively. (E) CPPA provides correlation analysis of pairwise proteins or phosphorylation sites, and query interface of similar expression pattern through “Correlation and similarity analysis” function. (F) CPPA provides survival analysis and query results about survival relevant proteins and phosphorylation sites.

General Analysis

Index of CPPA provides a simple query interface that takes Gene Symbol (e.g., SF3B1), UniProt ID (e.g., O75533), or Ensembl ID (e.g., ENSG00000115524) as input. After entering an identifier and clicking the “Search!” button, the website will navigate to pan-cancer analysis webpage, which presents the protein expression profile by box plot or violin plot and the distribution of phosphorylation sites through lollipop plot across all cancer types (Figure 2A). In addition, this module also provides basic information about proteins and external websites linking to this protein, including UniProt, GeneCard, PhosphoSitePlus, HPA, and so on.

Differential Profile

Dysregulation of protein expression or protein phosphorylation is one of the major causes of occurrence and development of cancer.25 Identifying differentially expression profiles between normal and tumor tissues is very helpful for inferring cancer driver factors, seeking specific biomarkers, and screening potential anticancer drug targets.26 CPPA provides an interface for exhibiting differential abundance profiles for proteins and protein phosphorylation. The significance of differences was calculated between normal and tumor samples by using the Wilcoxon signed-rank test, and then the Benjamini–Hochberg tutorial was applied to correct the P-values for better accuracy. This module allows users to obtain the differential expression profiles of tens of thousands of proteins and protein phosphorylation sites in normal and tumor samples according to user-specified P-value and fold change cutoffs (Figure 2B). For proteomic data, CPPA also provides functional enrichment analysis based on MsigDB aggregated gene sets, such as Hallmark, GO, KEGG, Reactome, WikiPathway, among others (Figure S1A). In addition, kinase-substrate enrichment analysis is added to CPPA for exploring kinase activity in cancer (Figure S1B).

Protein Phosphorylation Analysis

A protein can have multiple phosphorylation sites, which are likely to be different in abundance and molecular function. For example, depending on the sites involved, phosphorylation of the Elk-1 TAD by a single kinase, ERK, can either promote or inhibit mediator interaction, thereby modulating transcriptional activation.27 CPPA provides an interface that allows users to query all phosphorylation sites of a specific protein in a particular cancer type. The lollipop plot exhibits the distribution of phosphorylation sites on the protein, and the number of tissue samples in which these sites were identified. The heat map shows log2-transformed median abundance of phosphorylation sites across all normal and tumor tissues (Figure 2C). CPPA integrates relationships between kinases and substrates from PhosphoSitePlus and NetworKIN to inform users about potential kinases of the phosphorylation site (Figure S1C). In addition, CPPA also uses linear motif algorithm to predict kinases base on amino acid sequence surrounding the phosphorylation site through Persues (Figure S1D).

Custom Expression Analysis

CPPA provides a module to dynamically generate the abundance profile of a given protein or protein phosphorylation sites based on user-defined cancer type. The results are visualized by box plot along with text summarizing the detailed statistical results and the number of samples. In addition, CPPA also provides an interface for mapping expression abundance with pathological stage based on structured patient clinical annotations. Furthermore, comparison of different proteins or phosphorylation sites across multiple cancer types can be done in CPPA website, and the analysis results will be shown by heat map (Figure 2D).

Correlation and Similarity Analysis

CPPA provides a module for correlation analysis of paired proteins or phosphorylation sites, and results are shown by scatter plot. Users are free to choose the algorithms used to calculate the correlation, such as the Pearson, Spearman, and Kendall correlation analysis. Moreover, queries for similar expression pattern of proteins or phosphorylation sites are integrated into an interface with the same webpage structure. One of the applications is to seek for proteins or phosphorylation sites which are highly or inverse correlated with targets of interest in terms of abundance (Figure 2E).

Survival Analysis

CPPA can provide the survival curves of user-specified protein or phosphorylation site in a given cancer type, in which the Kaplan–Meier (KM) model is used to calculate the survival probability along with death event and survival time. Users can choose different stratification methods, such as Median, Tertiles, Quartile, and Best-cutoff, to classify samples into either high or low expression groups according to abundance. This module also calculates P-value through log-rank test and hazard ratio by cox proportional regression analysis. Besides, users can also choose whether or not to display the 95% confidence interval in survival curves. Similar to correlation analysis, CPPA also provides an interface for screening proteins or phosphorylation sites that are highly or inverse correlated with patient survival status in a given cancer type. The rank of correlation is sorted according to P-value and hazard ratio (Figure 2F).

Conclusion and Discussion

CPPA is a comprehensive and interactive web tool for mining characteristics of protein and protein phosphorylation sites from nearly 1500 patients spanning 13 different types of cancers, making use of proteomic and phosphoproteomic data from published data sets, particularly those from CPTAC. In order to increase reproducibility of the analysis results, CPPA standardizes the analysis pipeline through MaxQuant to reduce nonbiological variations. CPPA provides a variety of analysis results through multiple predesigned interactive web modules, including differential expression analysis, protein phosphorylation analysis, expression abundance analysis, correlation analysis, survival analysis, etc. Analysis results include approximately 17 000 proteins and more than 250 000 phosphorylation sites, covering the majority of the human proteome and phosphoproteome. Through interactive web operation, experimental biologists with basic bioinformatics skills can click a button to retrieve a series of analysis results and obtain images and tables for publication.

Because of the limitation of available large-scale high-resolution mass spectrometry-based data, proteomic and phosphoproteomic data collected by CPPA only covered 13 cancer types. With the development of cancer biology, CPPA will continuously collect more proteomic and phosphoproteomic data from other cancer types and be updated accordingly. Furthermore, PTMs is an important way to increase the functional diversity of proteins. In addition to phosphorylation, methylation, glycosylation, ubiquitination, acetylation, succinylation, other PTM types can almost affect all aspects of protein function, promoting or suppressing cancer development. We envision that CPPA can analyze more types of protein modifications with the development of various PTM omics, providing further insights into our understanding of cancer.

Acknowledgments

This work was supported by the Ministry of Science and Technology of China (2020YFA0112300, 2020YFA0803600), National Natural Science Foundation of China (82125028, 91953114, 31871319, 81761128015, and 81861130370), Natural Science Foundation of Fujian Province of China (2020J02004), and the Fundamental Research Funds for the Central University (20720190145 and 20720220003) to W. Liu. This work was also supported by the China Postdoctoral Science Foundation (2022M720119) to Zao-zao Zheng. We thank all other members from Liu’s lab for providing critical comments.

Data Availability Statement

Cancer Proteome and Phosphorylation Atlas (CPPA) analysis platform is available at https://cppa.site/cppa. This web tool is free and open to all users and there is no login requirement. The “Help” webpage of CPPA includes step-by-step instructions.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jproteome.2c00512.

  • Table S1: Description of data sets contained in CPPA (XLSX)

  • Figure S1: Examples of CPPA interactive results (PDF)

Author Contributions

W.L. and G.H. conceived the original ideas, designed the project, and wrote the manuscript. G.H. built the system’s back end. G.H. and Z.Z. designed the web interface. G.H. and Y.H. standardized the analysis pipeline of proteomic and phosphoproteomic raw data and constructed the SQLite data sets. D.W. tested and provided critical comments for the websites.

Author Contributions

§ G.H., Z.Z., Y.H., and D.W. contributed equally.

The authors declare no competing financial interest.

Supplementary Material

pr2c00512_si_001.xlsx (12.7KB, xlsx)
pr2c00512_si_002.pdf (3.8MB, pdf)

References

  1. International network of cancer genome projects. Nature 2010, 464, 993–998. 10.1038/nature08987. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Weinstein J. N.; et al. The Cancer Genome Atlas Pan-Cancer analysis project. Nat. Genet. 2013, 45, 1113–1120. 10.1038/ng.2764. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Yao J.; et al. Tumor subtype-specific cancer-testis antigens as potential biomarkers and immunotherapeutic targets for cancers. Cancer Immunol Res. 2014, 2, 371–379. 10.1158/2326-6066.CIR-13-0088. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Comprehensive molecular characterization of urothelial bladder carcinoma. Nature 2014, 507, 315–322. 10.1038/nature12965. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Rodriguez H.; Pennington S. R. Revolutionizing Precision Oncology through Collaborative Proteogenomics and Data Sharing. Cell 2018, 173, 535–539. 10.1016/j.cell.2018.04.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Edwards N. J.; et al. The CPTAC Data Portal: A Resource for Cancer Proteomics Research. J. Proteome Res. 2015, 14, 2707–2713. 10.1021/pr501254j. [DOI] [PubMed] [Google Scholar]
  7. Lin M. H.; et al. Benchmarking differential expression, imputation and quantification methods for proteomics data. Brief Bioinform 2022, 10.1093/bib/bbac138. [DOI] [PubMed] [Google Scholar]
  8. Liu W.; et al. Large-scale and high-resolution mass spectrometry-based proteomics profiling defines molecular subtypes of esophageal cancer for therapeutic targeting. Nat. Commun. 2021, 12, 4961. 10.1038/s41467-021-25202-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Krug K.; et al. Proteogenomic Landscape of Breast Cancer Tumorigenesis and Targeted Therapy. Cell 2020, 183, 1436–1456. 10.1016/j.cell.2020.10.036. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Vasaikar S.; et al. Proteogenomic Analysis of Human Colon Cancer Reveals New Therapeutic Opportunities. Cell 2019, 177, 1035–1049. 10.1016/j.cell.2019.03.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gao Q.; et al. Integrated Proteogenomic Characterization of HBV-Related Hepatocellular Carcinoma. Cell 2019, 179, 561–577. 10.1016/j.cell.2019.08.052. [DOI] [PubMed] [Google Scholar]
  12. Nørregaard Jensen O. Modification-specific proteomics: characterization of post-translational modifications by mass spectrometry. Curr. Opin Chem. Biol. 2004, 8, 33–41. 10.1016/j.cbpa.2003.12.009. [DOI] [PubMed] [Google Scholar]
  13. Derouiche A.; Cousin C.; Mijakovic I. Protein phosphorylation from the perspective of systems biology. Curr. Opin Biotechnol 2012, 23, 585–590. 10.1016/j.copbio.2011.11.008. [DOI] [PubMed] [Google Scholar]
  14. Cerami E.; et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov 2012, 2, 401–404. 10.1158/2159-8290.CD-12-0095. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Tang Z.; et al. GEPIA: a web server for cancer and normal gene expression profiling and interactive analyses. Nucleic Acids Res. 2017, 45, W98–w102. 10.1093/nar/gkx247. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Zhang Y.; Chen F.; Chandrashekar D. S.; Varambally S.; Creighton C. J. Proteogenomic characterization of 2002 human cancers reveals pan-cancer molecular subtypes and associated pathways. Nat. Commun. 2022, 13, 2669. 10.1038/s41467-022-30342-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Chandrashekar D. S.; et al. UALCAN: A Portal for Facilitating Tumor Subgroup Gene Expression and Survival Analyses. Neoplasia 2017, 19, 649–658. 10.1016/j.neo.2017.05.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. et al. The Genotype-Tissue Expression (GTEx) pilot analysis: multitissue gene regulation in humans. Science 2015, 348, 648–660. 10.1126/science.1262110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Moore H. F.; et al. The Genotype-Tissue Expression (GTEx) project. Nat. Genet. 2013, 45, 580–585. 10.1038/ng.2653. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Uhlén M.; et al. Proteomics. Tissue-based map of the human proteome. Science 2015, 347, 1260419. 10.1126/science.1260419. [DOI] [PubMed] [Google Scholar]
  21. Vasaikar S. V.; Straub P.; Wang J.; Zhang B. LinkedOmics: analyzing multi-omics data within and across 32 cancer types. Nucleic Acids Res. 2018, 46, D956–D963. 10.1093/nar/gkx1090. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Cox J.; Mann M. MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification. Nat. Biotechnol. 2008, 26, 1367–1372. 10.1038/nbt.1511. [DOI] [PubMed] [Google Scholar]
  23. Cox J.; et al. Andromeda: a peptide search engine integrated into the MaxQuant environment. J. Proteome Res. 2011, 10, 1794–1805. 10.1021/pr101065j. [DOI] [PubMed] [Google Scholar]
  24. Yu S. H.; Kyriakidou P.; Cox J. Isobaric Matching between Runs and Novel PSM-Level Normalization in MaxQuant Strongly Improve Reporter Ion-Based Quantification. J. Proteome Res. 2020, 19, 3945–3954. 10.1021/acs.jproteome.0c00209. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Lundby A.; et al. Oncogenic Mutations Rewire Signaling Pathways by Switching Protein Recruitment to Phosphotyrosine Sites. Cell 2019, 179, 543–560. 10.1016/j.cell.2019.09.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Noujaim J.; Payne L. S.; Judson I.; Jones R. L.; Huang P. H. Phosphoproteomics in translational research: a sarcoma perspective. Ann. Oncol 2016, 27, 787–794. 10.1093/annonc/mdw030. [DOI] [PubMed] [Google Scholar]
  27. Mylona A.; et al. Opposing effects of Elk-1 multisite phosphorylation shape its response to ERK activation. Science 2016, 354, 233–237. 10.1126/science.aad1872. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

pr2c00512_si_001.xlsx (12.7KB, xlsx)
pr2c00512_si_002.pdf (3.8MB, pdf)

Data Availability Statement

Cancer Proteome and Phosphorylation Atlas (CPPA) analysis platform is available at https://cppa.site/cppa. This web tool is free and open to all users and there is no login requirement. The “Help” webpage of CPPA includes step-by-step instructions.

Articles from Journal of Proteome Research are provided here courtesy of American Chemical Society

RESOURCES