Satwik Rajaram

NeatMap - non-clustering heat map alternatives in R

BackgroundThe clustered heat map is the most popular means of visualizing genomic data. It compactly displays a large amount of data in an intuitive format that facilitates the detection of hidden structures and relations in the data. However, it is hampered by its use of cluster analysis which does not always respect the intrinsic relations in the data, often requiring non-standardized reordering of rows/columns to be performed post-clustering. This sometimes leads to uninformative and/or misleading conclusions. Often it is more informative to use dimension-reduction algorithms (such as Principal Component Analysis and Multi-Dimensional Scaling) which respect the topology inherent in the data. Yet, despite their proven utility in the analysis of biological data, they are not as widely used. This is at least partially due to the lack of user-friendly visualization methods with the visceral impact of the heat map.ResultsNeatMap is an R package designed to meet this need. NeatMap offers a variety of novel plots (in 2 and 3 dimensions) to be used in conjunction with these dimension-reduction techniques. Like the heat map, but unlike traditional displays of such results, it allows the entire dataset to be displayed while visualizing relations between elements. It also allows superimposition of cluster analysis results for mutual validation. NeatMap is shown to be more informative than the traditional heat map with the help of two well-known microarray datasets.ConclusionsNeatMap thus preserves many of the strengths of the clustered heat map while addressing some of its deficiencies. It is hoped that NeatMap will spur the adoption of non-clustering dimension-reduction algorithms.

PhenoRipper: software for rapidly profiling microscopy images.

Supplementary Figure 1 PhenoRipper flow chart. Supplementary Figure 2 Comparison of classification performance for various unsupervised image analysis methods. Supplementary Figure 3 PhenoRipper enables intuitive exploration of phenotypic space spanned by cellular perturbations. Supplementary Figure 4 Parameter dependence of PhenoRipper results. Supplementary Figure 5 PhenoRipper identifies meaningful features from hard-to-segment data. Supplementary Figure 6 PhenoRipper can rapidly analyze large datasets and provide biologically interpretable grouping. Supplementary Methods

SimuCell: a flexible framework for creating synthetic microscopy images

1. Swedlow, J.R. Nat. Cell Biol. 13, 183 (2011). 2. Faloutsos, C. et al. J. Intell. Inf. Syst. 3, 231–262 (1994). 3. Allan, C. et al. Nat. Methods 9, 245–253 (2012). 4. Glory, E. & Murphy, R.F. Dev. Cell 12, 7–16 (2007). 5. Wu, L., Faloutsos, C., Sycara, K.P. & Payne, T.R. in Proc. 26th Int. Conf. Very Large Data Bases (eds., Abbadi, A.E. et al.) 297–306 (Morgan Kaufmann, 2000). 6. Huang, K., Lin, J., Gajnak, J.A. & Murphy, R.F. in Proc. 2002 IEEE Int. Symp. Biomed. Imaging, 325–328 (2002).

On comparing heterogeneity across biomarkers.

Microscopy reveals complex patterns of cellular heterogeneity that can be biologically informative. However, a limitation of microscopy is that only a small number of biomarkers can typically be monitored simultaneously. Thus, a natural question is whether additional biomarkers provide a deeper characterization of the distribution of cellular states in a population. How much information about a cells phenotypic state in one biomarker is gained by knowing its state in another biomarker? Here, we describe a framework for comparing phenotypic states across biomarkers. Our approach overcomes the current limitation of microscopy by not requiring costaining biomarkers on the same cells; instead, we require staining of biomarkers (possibly separately) on a common collection of phenotypically diverse cell lines. We evaluate our approach on two image datasets: 33 oncogenically diverse lung cancer cell lines stained with 7 biomarkers, and 49 less diverse subclones of one lung cancer cell line stained with 12 biomarkers. We first validate our method by comparing it to the “gold standard” of costaining. We then apply our approach to all pairs of biomarkers and use it to identify biomarkers that yield similar patterns of heterogeneity. The results presented in this work suggest that many biomarkers provide redundant information about heterogeneity. Thus, our approach provides a practical guide for selecting independently informative biomarkers and, more generally, will yield insights into both the connectivity of biological networks and the complexity of the state space of biological systems.

A simple image correction method for high-throughput microscopy

Advances in high-throughput fluorescence microscopy (HTFM) and image informatics have allowed heterogeneous single-cell phenotypes to be characterized and related to putative biological functions. However, it is underappreciated that accurate estimates of single-cell phenotypic variability depend on image correction. Further, the methods used throughout the HTFM literature do not generally take advantage of shared image properties, instead correcting each image independently. In this brief note, we first recall the basic mathematical framework underlying image correction. We then show how the true distributions of commonly measured single-cell features are altered in uncorrected images. Finally, we report an observation that allows the use of a simple method for accurate image correction in HTFM. The intensity of a pixel, I, can be modeled by I = S(F + B), where: S is shading due to uneven illumination, F is the biologically relevant foreground fluorescence, and B is the background fluorescence1–3 (see Supplemental §1 for full model description). Image correction is a data preprocessing step whose goal is to obtain F. (Subsequent normalization steps, not discussed here, may be applied to deal with systematic experimental errors, such as row or column biases in microtiter plates4,5.) How is apparent cellular variability (e.g. standard deviation or coefficient of variation) affected when S or B are not removed from images? In Figure 1a and Supplemental §2, we analyze such effects for the distributions of commonly used single-cell features, namely: Average, Total or Ratiometric biomarker intensities. For these simple features, background and shading can alter apparent phenotypic variability in surprising ways. For example, the effects of either on the Ratio feature are generally unpredictable, while background increases Total feature variability due to underlying differences in cell size. Such changes to single-cell feature distributions can be particularly important in screening, where variation is commonly used to determine whether experimental conditions differ statistically from controls. Figure 1 a, Image background and shading can alter measured distributions of cellular phenotypes. The effects of image background (“B”, blue) or shading (“S”, red) on distributions of three measured cellular features were mathematically ... In practice, the challenge for correction is to estimate shading within an image, as subsequent background subtraction is relatively simple. How then should shading in large HTFM image datasets be estimated? In principle, every image from a microtiter plate could have its own unique shading pattern and background. Indeed, this is an (implicit) assumption of correction methods common to HTFM studies. However, we find that shading is not unique to every image, but is rather predictable based on image position within a well (Figures 1b, S1, S2). This observation suggests an image correction strategy for HTFM that only depends on within-well position. In short, images are corrected by within-well position using reference shading patterns estimated from the data (Figure 1b, Supplemental §3). This empirical approach using shared image properties across well positions is simpler than commonly used methods that estimate correction parameters for each individual image, yet more accurate than applying one set of correction parameters to all images. Accurate estimates of variability are important for large-scale studies of single-cell phenotypes, which are increasingly used to link functional significance to patterns of cellular heterogeneity.

Combination Therapy Targeting BCL6 and Phospho-STAT3 Defeats Intratumor Heterogeneity in a Subset of Non-Small Cell Lung Cancers

Oncogene-specific changes in cellular signaling have been widely observed in lung cancer. Here, we investigated how these alterations could affect signaling heterogeneity and suggest novel therapeutic strategies. We compared signaling changes across six human bronchial epithelial cell (HBEC) strains that were systematically transformed with various combinations of TP53, KRAS, and MYC-oncogenic alterations commonly found in non-small cell lung cancer (NSCLC). We interrogated at single-cell resolution how these alterations could affect classic readouts (β-CATENIN, SMAD2/3, phospho-STAT3, P65, FOXO1, and phospho-ERK1/2) of key pathways commonly affected in NSCLC. All three oncogenic alterations were required concurrently to observe significant signaling changes, and significant heterogeneity arose in this condition. Unexpectedly, we found two mutually exclusive altered subpopulations: one with STAT3 upregulation and another with SMAD2/3 downregulation. Treatment with a STAT3 inhibitor eliminated the upregulated STAT3 subpopulation, but left a large surviving subpopulation with downregulated SMAD2/3. A bioinformatics search identified BCL6, a gene downstream of SMAD2/3, as a novel pharmacologically accessible target of our transformed HBECs. Combination treatment with STAT3 and BCL6 inhibitors across a panel of NSCLC cell lines and in xenografted tumors significantly reduced tumor cell growth. We conclude that BCL6 is a new therapeutic target in NSCLC and combination therapy that targets multiple vulnerabilities (STAT3 and BCL6) downstream of common oncogenes, and tumor suppressors may provide a potent way to defeat intratumor heterogeneity. Cancer Res; 77(11); 3070-81. ©2017 AACR.

Sampling strategies to capture single-cell heterogeneity

Advances in single-cell technologies have highlighted the prevalence and biological significance of cellular heterogeneity. A critical question researchers face is how to design experiments that faithfully capture the true range of heterogeneity from samples of cellular populations. Here we develop a data-driven approach, illustrated in the context of image data, that estimates the sampling depth required for prospective investigations of single-cell heterogeneity from an existing collection of samples.

Rapid Analysis and Exploration of Fluorescence Microscopy Images

Despite rapid advances in high-throughput microscopy, quantitative image-based assays still pose significant challenges. While a variety of specialized image analysis tools are available, most traditional image-analysis-based workflows have steep learning curves (for fine tuning of analysis parameters) and result in long turnaround times between imaging and analysis. In particular, cell segmentation, the process of identifying individual cells in an image, is a major bottleneck in this regard. Here we present an alternate, cell-segmentation-free workflow based on PhenoRipper, an open-source software platform designed for the rapid analysis and exploration of microscopy images. The pipeline presented here is optimized for immunofluorescence microscopy images of cell cultures and requires minimal user intervention. Within half an hour, PhenoRipper can analyze data from a typical 96-well experiment and generate image profiles. Users can then visually explore their data, perform quality control on their experiment, ensure response to perturbations and check reproducibility of replicates. This facilitates a rapid feedback cycle between analysis and experiment, which is crucial during assay optimization. This protocol is useful not just as a first pass analysis for quality control, but also may be used as an end-to-end solution, especially for screening. The workflow described here scales to large data sets such as those generated by high-throughput screens, and has been shown to group experimental conditions by phenotype accurately over a wide range of biological systems. The PhenoBrowser interface provides an intuitive framework to explore the phenotypic space and relate image properties to biological annotations. Taken together, the protocol described here will lower the barriers to adopting quantitative analysis of image based screens.

Abstract 3950: A novel combination therapy targeting BCL6 and phospho-STAT3 defeats intratumor heterogeneity in a subset of non-small cell lung cancers

Purpose: Lung cancer is a disease of great oncogenotype complexity (oncogenes and tumor suppressor gene alterations). These alterations can appear in different combinations even within histologically defined lung cancer subtypes. The success of targeted therapy has led to a search for oncogenotype-specific therapies. But, no one therapy fits all oncogenotypes. Here, we investigate whether characterization of oncogene-specific alterations in cellular signaling at single cell level indicate heterogeneity even within cells from the same patient with defined oncogenotype, and whether they can suggest new targets to deal with this heterogeneity. Methods: We compared signaling alterations in single cells for β-CATENIN, SMAD2/3, phospho-STAT3, P65, FOXO1 and phospho-ERK1/2 among a collection of Human Bronchial Epithelial Cells (HBECs) that have been oncogenically transformed with combinations of TP53, K-RAS, and MYC, commonly found alterations in non-small cell lung cancer (NSCLC). We studied ~3000 cells/signaling marker/HBEC oncogenotype variant using immunofluorescence assays and single-cell image analysis (>1M data points). For downstream target identification and validation we utilized gene expression, Western blot and siRNA mediated knockdown assays. We utilized inhibitors to STAT3 and BCL6 in MTT drug sensitivity and colony formation assay in a panel of NSCLC lines. We used xenografted subcutaneous tumors for the in vivo validation of our results. Results: When all three oncogenic changes were present and the HBECs were tumorigenic, we observed STAT3 upregulation and SMAD2/3 downregulation. Interestingly, these STAT3 and SMAD2/3 signaling changes were found to be mutually exclusive in single cells within the transformed HBEC strain. We targeted the STAT3 upregulated subpopulation with the STAT3 inhibitor Stattic. But, Stattic treatment failed to eliminate the SMAD2/3 downregulated subpopulation. To target the SMAD2/3 down-regulated subpopulation, we identified BCL6, a downstream gene of SMAD2/3, as a novel target in transformed HBECs. Next, to test the generality of BCL6 as a target, we studied 5 NSCLC cell lines with various level of BCL6 expression: H1693, H1819, H1993, HCC827 and H2009. Our data suggests that BCL6 can also be a therapeutic target in a subset of NSCLC lines. Then we tested the response of these NSCLC lines to a combination of BBI608 (potent STAT3 inhibitor) and FX1 (BCL6 inhibitor). The combination treatment eliminated more cancer cells than the single treatments alone. Finally, we confirmed the benefit of the combination therapy in H1993 xenografted tumors. Conclusions: We conclude that BCL6 is a new therapeutic target in NSCLC and combination therapy that targets multiple vulnerabilities (Phospho-STAT3 and BCL6) downstream of common oncogenes and tumor suppressors (TP53, K-RAS, and MYC) may provide a potent way to defeat intra-tumor heterogeneity. Citation Format: Dhruba Deb, Satwik Rajaram, Jill E. Larsen, Patrick P. Dospoy, Rossella Marullo, Longshan Li, Kimberley Avila, Leandro Cerchietti, John D. Minna, Lani F. Wu, Steven J. Altschuler. A novel combination therapy targeting BCL6 and phospho-STAT3 defeats intratumor heterogeneity in a subset of non-small cell lung cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3950. doi:10.1158/1538-7445.AM2017-3950