image: This pie chart illustrates the distribution of visualization tools in the FigureYa resource package across three dimensions: research type (outer ring), analysis method (middle ring), and output format (inner ring). Major research categories include single‐cell analysis (pink), machine learning (yellow), genomic analysis (green), immunological analysis (light blue), expression profiling (dark blue), and survival analysis (purple). Each segment displays the proportion of tools within each category. This multi‐dimensional classification system enables researchers to efficiently identify appropriate tools based on specific research objectives and data characteristics.
Credit: t: Xiaofan Lu, Kailai Li, Zongcheng Li, Anqi Lin, Long Zhao, Rongfang Shen, Zhougeng Xu, Jianing Gao, Dekang Lv, Yasi Zhang, Taojun Ye, Junyi Shen, Yijing Chen, Hui Huang, Zhaodong Hao, Dongqiang Zeng, Haitao Wang, Shipeng Guo, Wen Wang, Yi Xiong, Yin Li, Hao Li, Jingze Gao, Qian Liu, Bin Wei, Jiawei Shi, Shuwen Cheng, Guoqi Li, Yuan Tang, Enyu Lin, Houshi Xu, Guoshuang Feng, Yangzhige He, Yu Sun, Xiaojian Liu, Yufang Wang, Wenxin Song, Jinen Song, Saisai Tian, Ya Zhang, Jie Zhang, Zhongtian Xu, Chengli Song, Yingying Zhang, Hao Wu, Chunhui Gao, Erqiang Hu, Chen Yang, Jiacheng Lou, Di Wang, Xuanyu Wang, Peng Luo, Guangchuang Yu, Ying Ge
With the widespread application of technologies such as high-throughput sequencing, large-scale clinical trials, and complex computational simulations, the amount of data that researchers must process continues to grow exponentially. Medical researchers face significant challenges in integrating and visualizing large-scale multidimensional data. In multi-omics research, it is necessary to analyze multiple data sources, such as genomes and transcriptomes, simultaneously; however, existing tools are often limited to a single data type. Professional visualization tools, such as R's ggplot2, require advanced programming skills, whereas user-friendly tools like Excel offer only limited functionality. Even simple visualization tasks often require substantial coding, creating barriers for researchers without a computational background. These tool limitations significantly reduce research efficiency. FigureYa is a modular R-based visualization framework. It introduces the innovative concept of “ready-to-use visual code and sample data integration,” effectively eliminating technical barriers to scientific visualization through well-designed pre-configured code templates and accompanying sample datasets. Its application scenarios span key stages of the research process, including figure preparation for publications, conference presentations, and exploratory data analysis. Users simply need to format their data similarly to the provided examples to generate professional-grade visualizations. Each script includes bilingual annotations, usage guidance, example datasets, and pre-rendered outputs. The GitHub repository is organized as a navigable structure akin to a web interface, allowing limited programming users to follow steps with minimal technical background.
The FigureYa code resource package comprises 317 highly specialized visualization scripts, covering major data types and analytical scenarios in biomedical research. By research type, gene expression profiling, immunophenotyping, survival analysis, and single-cell analysis represent the most prevalent categories. In terms of analysis type, enrichment analysis, differential analysis, and correlation analysis are the most frequently used. With respect to output type, heatmaps, line charts, and scatter plots are the most commonly utilized visualization formats.
The scope of the FigureYa code resource package reflects the design philosophy of “from fundamentals to cutting-edge, from single to integrated.” In terms of basic statistical visualization, the package includes standard descriptive charts—such as FigureYa12box and FigureYa59volcano—which are optimized to automatically handle significance markers and multi-group comparisons. For domain-specific visualization, the resource package covers essential chart types across major biomedical research areas, including genomics, transcriptomics, and clinical research. For emerging technologies, FigureYa offers end-to-end visualization tools for single-cell analysis, spatial transcriptomics (e.g., FigureYa239ST_PDAC), and multi-omics integration (e.g., FigureYa258SNF). Additionally, the package includes specialized tools for drug sensitivity analysis, immune microenvironment profiling, and epigenetics research. This comprehensive scope ensures that researchers can identify appropriate visualization tools regardless of their research stage or disciplinary focus.
The FigureYa code resource package establishes an interconnected and composable visualization ecosystem. This connectivity is manifested at three levels: data flow integration, analytical workflow linkage, and compositional visualization integration. At the data flow level, the resource package includes data preprocessing and format conversion tools—such as FigureYa21TCGA2table and FigureYa22FPKM2TPM—forming a complete data pipeline. At the analysis workflow level, multiple scripts can be combined to construct a complete analytical pipeline. For example, in transcriptomic differential expression analysis, FigureYa119Multiclasslimma, FigureYa9heatmap, FigureYa59volcano, and FigureYa60GSEA_clusterProfiler can be used sequentially to create a complete pipeline from differential expression testing to functional interpretation. At the visualization integration level, multiple visualization scripts can be combined to generate complex multi-panel figures. For instance, FigureYa69cancerSubtype integrates clustering heatmaps, survival analysis, and clinical correlation analysis into a unified visual framework.
The four principal categories of visualization techniques implemented in FigureYa: basic statistical visualization (e.g., FigureYa12box, FigureYa126CorrelationHeatmap and FigureYa76corrgram), survival and clinical data visualization (e.g., FigureYa1survivalCurve, FigureYa30nomogram and FigureYa193RiskTable), omics data visualization (e.g., FigureYa3genomeView, FigureYa60GSEA_clusterProfiler and FigureYa122mut2expr), and cutting-edge visualization techniques (e.g., FigureYa224scMarker, FigureYa239ST_PDAC and FigureYa293machineLearning). Basic statistical visualization include box plot and violin plot series, correlation analysis visualization and parametric customization capabilities; survival analysis and clinical data visualization encompass survival curve series, clinical prediction model visualization and presentation of multivariate analysis results; omics data visualizations cover genomic and epigenomic visualization, transcriptome data analysis graphics and multi-omics integration visualization; and visualization of cutting-edge technologies include visualization of single-cell analysis, spatial transcriptomics data visualization and visualization of machine learning results.
FigureYa has transformed biomedical data visualization through its “plug-and-play” workflow and standardized code architecture. The workflow developed in this study streamlines the complex visualization process into four coherent steps: selecting an appropriate code template, substituting user-specific data, executing standardized scripts, and generating publication-quality chart outputs. This design enables researchers without programming expertise to efficiently produce visualization results that meet publication standards. The standardized code architecture follows a six-stage structure encompassing data input, cleaning, transformation, analysis, visualization generation, and result saving, with detailed annotations and parameter descriptions at each stage to ensure code readability and maintainability.
Multi-dimensional evaluation results indicate that FigureYa achieves an optimal balance across six key dimensions: automation level, reproducibility, professional standards, time efficiency, ease of learning, and moderate customizability. This innovative workflow not only removes technical barriers but, more importantly, enables researchers to focus on scientific inquiry rather than on technical implementation details.
COI Statement
Peng Luo is the Editorial Board members of iMetaMed and co-authors of this article. He was excluded from editorial decision-making related to the acceptance of this article for publication in the journal.