Integrative Analysis of Histopathological Images and Genomic Data Predicts Clear Cell Renal Cell Carcinoma Prognosis

Cheng, Jun; Zhang, Jie; Ying, Han; Wang, Xusheng; Ye, Xiufen; Meng, Yuebo; Parwani, Anil V.; Han, Zhi; Feng, Qianjin; Huang, Kun

doi:10.1158/0008-5472.can-17-0313

Cited by 119 publications

(116 citation statements)

References 51 publications

Supporting

Mentioning

116

Contrasting

Order By: Relevance

“…This has been applied in the genomics field before Cheng et al, 2017;Agrahari et al, 2018), but to the best of our knowledge, has not been explored in the proteomics field thus far. By combining protein expression with protein interaction information, our pipeline generates so-called 'cliques', which are subsequently used to recluster the single cells on an embedding of choice (UMAP or tSNE).…”

Section: Eigenprotein Analysis Of Scms Datamentioning

confidence: 99%

Quantitative Single-Cell Proteomics as a Tool to Characterize Cellular Hierarchies

Schoof

Furtwängler

Üresin

et al. 2019

Preprint

101

View full text Add to dashboard Cite

Key words: acute myeloid leukemia / computational biology / proteomics / single-cell approaches / tandem mass tags (TMT)Final character count: 47,782 AbstractIn recent years, cellular life science research has experienced a significant shift, moving away from conducting bulk cell interrogation towards single-cell analysis. It is only through single cell analysis that a complete understanding of cellular heterogeneity, and the interplay between various cell types that are fundamental to specific biological phenotypes, can be achieved. Single-cell assays at the protein level have been predominantly limited to targeted, antibody-based methods. However, here we present an experimental and computational pipeline, which establishes a comprehensive single-cell mass spectrometry-based proteomics workflow.By exploiting a leukemia culture system, containing functionally-defined leukemic stem cells, progenitors and terminally differentiated blasts, we demonstrate that our workflow is able to explore the cellular heterogeneity within this aberrant developmental hierarchy. We show our approach is capable to quantifying hundreds of proteins across hundreds of single cells using limited instrument time. Furthermore, we developed a computational pipeline (SCeptre), that effectively clusters the data and permits the extraction of cell-specific proteins and functional pathways. This proof-of-concept work lays the foundation for future global single-cell proteomics studies. Duployez et al, 2019;Ng et al, 2016), their accuracy has proven limitation when used as a proxy for protein levels (Vogel & Marcotte, 2012; Khan et al, 2013). Therefore, to gain a thorough understanding of what occurs in a cell at the protein level, on a global scale, MSbased approaches are the sole way to accomplish this. Being the cellular workhorses, there is much knowledge to be gained from mechanisms occurring at the protein level, either through enzyme activity, post-translational modifications or protein degradation/proteolysis; hence the great need for protein level approaches at the single-cell level.A few years ago, a novel type of flow cytometry was established; by combining traditional flow cytometry workflows with mass spectrometry, a new analysis method termed Mass Cytometry was developed, more commonly referred to as CyTOF (Newell et al, 2012;Bodenmiller et al, 2012). This allows the simultaneous readout of tens of markers simultaneously, allowing single-cell analysis of pre-defined sets of proteins or posttranslational modifications (PTMs). This method, however, relies heavily on previously

show abstract

Section: Eigenprotein Analysis Of Scms Datamentioning

confidence: 99%

Quantitative Single-Cell Proteomics as a Tool to Characterize Cellular Hierarchies

Schoof

Furtwängler

Üresin

et al. 2019

Preprint

101

View full text Add to dashboard Cite

show abstract

“…We estimated the risk index of each patient based on each morphological feature. A two level cross validation strategy was used to validate our model 23 .…”

Section: Mapping High Probability Regions and Survival Analysismentioning

confidence: 99%

“…Cheng et.al. (2017) 23 extracted image features from TCGA tumor slides to develop a lasso-regularized Cox model for predicting survival of KIRC patients.…”

Section: Introductionmentioning

confidence: 99%

A Deep learning approach for Pan-Renal Cell Carcinoma classification and survival prediction from histopathology images

Tabibu

Vinod²,

Jawahar

2019

Preprint

View full text Add to dashboard Cite

Histopathological images contain morphological markers of disease progression that have diagnostic and predictive values. However, complex morphological information remains unutilized in unaided approach to histopathology. In this study, we demonstrate how deep learning framework can be used for an automatic classification of Renal Cell Carcinoma (RCC) subtypes, and for identification of features that predict survival outcome from digital histopathological images. Convolutional neural networks (CNN's) trained on whole-slide images distinguish clear cell and chromophobe RCC from normal tissue with a classification accuracy of 93.39 % and 87.34 %, respectively. Further, a CNN trained to distinguish clear cell, chromophobe and papillary RCC achieves a classification accuracy of 92.61 %. Here, we introduced a novel support vector machine based method to deal with data imbalance in multi-class classification to improve the accuracy. Finally, we extracted the morphological features from high probability tumor regions identified by the CNN to predict patient survival outcome of most common clear cell RCC. The generated risk index based on both tumor shape and nuclei features are significantly associated with patient survival outcome. These results highlight that deep learning can play a role in both cancer diagnosis and prognosis.

show abstract

“…Currently, the prognostic biomarkers of ccRCC mainly include tumor stage (7), UISS score (6), SSIGN score (8), and Leibovich score (9). However, the predictive accuracy of these biomarkers is limited and dependent on a pathologist's experience (10)(11)(12)(13). Therefore, reliable and powerful prognostic biomarkers of ccRCC are needed.…”

Section: Introductionmentioning

confidence: 99%

“…Aside from images, molecular characteristics, such as gene expression profiling, are widely adopted in predicting the clinical outcomes of cancers (23)(24)(25). Several studies have been conducted to predict ccRCC prognosis based on hand-crafted features (10) and single-module data (25,26). However, handcrafted feature extraction is tedious and dependent on the experience of researchers, and different module data can complement and reinforce one another in the prediction of ccRCC prognosis.…”

Section: Introductionmentioning

confidence: 99%

Integrative framework of cross-module deep biomarker for the prognosis of clear cell renal cell carcinoma

Ning

Pan

Xiao

et al. 2019

Preprint

View full text Add to dashboard Cite

Purpose:We aimed to integrate cross-module data for predicting the prognosis of clear cell renal cell carcinoma (ccRCC) based on deep learning and to explore the relationship between deep features from images and eigengenes form gene data. Experimental design:A total of 209 patients with ccRCC with computed tomography (CT), histopathological images and RNA sequences were enrolled. A deep biomarker-based integrative framework was proposed to construct a prognostic model. Deep features extracted from CT and histopathological images by using deep learning combined with eigengenes generated from functional genomic data were used to predict ccRCC prognosis. Furthermore, the relationship between deep features and eigengenes was explored, and two survival subgroups identified by integrative cross-module biomarkers were subjected to functional analysis. Results: The model based on the integrative framework stratified two subgroups of patients with a significant prognostic difference (P = 6.51e-6, concordance index [C-index] = 0.808, 95% confidence interval [CI] = 0.728-0.888) and outperformed the prediction based on their individual biomarkers in the independent validation cohort (n = 70, gene data: C-index = 0.452, CI = 0.336-0.567; histopathological images: C-index = 0.677, CI = 0.577-0.776; CT images: Cindex = 0.774, CI = 0.670-0.879). On the basis of statistical relationship, deep features correlated or complemented with eigengenes both enhanced the predictive performance of eigengenes (P = 0.439, correlated: C-index = 0.785, CI = 0.685-0.886; complemented: C-index = 0.778, CI = 0.683-0.872). The functional analysis of subgroups also exhibited reasonable results. 4 Conclusion:The model based on the integrative framework of cross-module deep biomarkers can efficiently predict ccRCC prognosis, and the framework with a code is shared to act as a reliable and powerful tool for further studies.

show abstract

Integrative Analysis of Histopathological Images and Genomic Data Predicts Clear Cell Renal Cell Carcinoma Prognosis

Cited by 119 publications

References 51 publications

Quantitative Single-Cell Proteomics as a Tool to Characterize Cellular Hierarchies

Quantitative Single-Cell Proteomics as a Tool to Characterize Cellular Hierarchies

A Deep learning approach for Pan-Renal Cell Carcinoma classification and survival prediction from histopathology images

Integrative framework of cross-module deep biomarker for the prognosis of clear cell renal cell carcinoma

Contact Info

Product

Resources

About