Limits...
A novel phenotypic dissimilarity method for image-based high-throughput screens.

Zhang X, Boutros M - BMC Bioinformatics (2013)

Bottom Line: Here we propose a new analysis method, PhenoDissim, which computes the phenotypic dissimilarity between cell populations via Support Vector Machine classification and cross validation.PhenoDissim is a novel analysis method for image-based high-throughput screen, relying on two parameters which can be automatically optimized without a priori knowledge.PhenoDissim is freely available as an R package.

View Article: PubMed Central - HTML - PubMed

Affiliation: German Cancer Research Center (DKFZ), Div, Signaling and Functional Genomics and Department of Cell and Molecular Biology, Medical Faculty Mannheim, Im Neuenheimer Feld 580, D-69120 Heidelberg, Germany. xianzhang@gmail.com.

ABSTRACT

Background: Discovering functional relationships of genes through cell-based phenotyping has become an important approach in functional genomics. High-throughput imaging offers the ability to quantitatively assess complex phenotypes after perturbation by RNA interference (RNAi). Such image-based high-throughput RNAi screening studies have facilitated the discovery of novel components of gene networks and their interactions. Images generated by automated microscopy are typically analyzed by extracting quantitative features of individual cells, resulting in large multidimensional data sets. Robust and sensitive methods to interpret these data sets and to derive biologically relevant information in a high-throughput and unbiased manner remain to be developed.

Results: Here we propose a new analysis method, PhenoDissim, which computes the phenotypic dissimilarity between cell populations via Support Vector Machine classification and cross validation. Applying this method to a kinome RNAi screening data set, we demonstrate that the proposed method shows a good replicate reproducibility, separation of controls and clustering quality, and we are able to identify siRNA phenotypes and discover potential functional links between genes.

Conclusions: PhenoDissim is a novel analysis method for image-based high-throughput screen, relying on two parameters which can be automatically optimized without a priori knowledge. PhenoDissim is freely available as an R package.

Show MeSH

Related in: MedlinePlus

Workflow for image-based screening, image quantification and phenotypic dissimilarity measure with SVM classification accuracy.A) Cells are seeded into 384-well plates and treated with siRNA by reverse transfection. After incubation for 48 hours, cells are fixed, permeabilized and immunostained for DNA, tubulin and actin and imaged with an automated microscope. B) Cell images are processed with nucleus and cell segmentation using the R packages EBImage and imageHTS. Each cell is represented by a 46 image-based feature vector. Every treatment generates a data matrix X[m,n], where m is the number of cells and n is the number of features. C) For each pair of RNAi treatments, SVM classification is performed on the virtually pooled cell population based on cell features. Classification accuracy is estimated by cross validation, and defined as the phenotypic dissimilarity between treatments.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4225524&req=5

Figure 1: Workflow for image-based screening, image quantification and phenotypic dissimilarity measure with SVM classification accuracy.A) Cells are seeded into 384-well plates and treated with siRNA by reverse transfection. After incubation for 48 hours, cells are fixed, permeabilized and immunostained for DNA, tubulin and actin and imaged with an automated microscope. B) Cell images are processed with nucleus and cell segmentation using the R packages EBImage and imageHTS. Each cell is represented by a 46 image-based feature vector. Every treatment generates a data matrix X[m,n], where m is the number of cells and n is the number of features. C) For each pair of RNAi treatments, SVM classification is performed on the virtually pooled cell population based on cell features. Classification accuracy is estimated by cross validation, and defined as the phenotypic dissimilarity between treatments.

Mentions: To understand phenotypes and their regulations, it is important to identify key genetic components as well as how they interact. Cell-based screening approaches have been successfully used to monitor the effect of individual gene knockdowns or small molecule treatments, identify key regulators contributing to the assessed phenotype and investigate their interactions [1,2]. Such high-throughput screening experiments can be divided into two categories: homogeneous intensity-based methods, such as reporter gene or cell viability assays, and image-based phenotyping approaches. Intensity-based methods usually report the average of cell populations, leading to scalar (or low dimensional) values per perturbation. Such screens have been designed, for example, to identify novel signaling pathway components by associating an intensity readout (e.g., luminescence or fluorescence) with a perturbation of a specific reporter gene activity [3-8]. In contrast, image-based methods mark cells with fluorescent dyes, and produce high-dimensional data sets based on images of phenotypes on a single cell level and consequently on cell populations [9-15]. Cellular phenotyping by imaging offers many advantages including flexible marker choices, subcellular resolution and ability to address cell population heterogeneity (Figure 1A), but also pose new challenges such as lower throughput, more complex infrastructure, and in particular, challenges in data analysis [16].


A novel phenotypic dissimilarity method for image-based high-throughput screens.

Zhang X, Boutros M - BMC Bioinformatics (2013)

Workflow for image-based screening, image quantification and phenotypic dissimilarity measure with SVM classification accuracy.A) Cells are seeded into 384-well plates and treated with siRNA by reverse transfection. After incubation for 48 hours, cells are fixed, permeabilized and immunostained for DNA, tubulin and actin and imaged with an automated microscope. B) Cell images are processed with nucleus and cell segmentation using the R packages EBImage and imageHTS. Each cell is represented by a 46 image-based feature vector. Every treatment generates a data matrix X[m,n], where m is the number of cells and n is the number of features. C) For each pair of RNAi treatments, SVM classification is performed on the virtually pooled cell population based on cell features. Classification accuracy is estimated by cross validation, and defined as the phenotypic dissimilarity between treatments.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4225524&req=5

Figure 1: Workflow for image-based screening, image quantification and phenotypic dissimilarity measure with SVM classification accuracy.A) Cells are seeded into 384-well plates and treated with siRNA by reverse transfection. After incubation for 48 hours, cells are fixed, permeabilized and immunostained for DNA, tubulin and actin and imaged with an automated microscope. B) Cell images are processed with nucleus and cell segmentation using the R packages EBImage and imageHTS. Each cell is represented by a 46 image-based feature vector. Every treatment generates a data matrix X[m,n], where m is the number of cells and n is the number of features. C) For each pair of RNAi treatments, SVM classification is performed on the virtually pooled cell population based on cell features. Classification accuracy is estimated by cross validation, and defined as the phenotypic dissimilarity between treatments.
Mentions: To understand phenotypes and their regulations, it is important to identify key genetic components as well as how they interact. Cell-based screening approaches have been successfully used to monitor the effect of individual gene knockdowns or small molecule treatments, identify key regulators contributing to the assessed phenotype and investigate their interactions [1,2]. Such high-throughput screening experiments can be divided into two categories: homogeneous intensity-based methods, such as reporter gene or cell viability assays, and image-based phenotyping approaches. Intensity-based methods usually report the average of cell populations, leading to scalar (or low dimensional) values per perturbation. Such screens have been designed, for example, to identify novel signaling pathway components by associating an intensity readout (e.g., luminescence or fluorescence) with a perturbation of a specific reporter gene activity [3-8]. In contrast, image-based methods mark cells with fluorescent dyes, and produce high-dimensional data sets based on images of phenotypes on a single cell level and consequently on cell populations [9-15]. Cellular phenotyping by imaging offers many advantages including flexible marker choices, subcellular resolution and ability to address cell population heterogeneity (Figure 1A), but also pose new challenges such as lower throughput, more complex infrastructure, and in particular, challenges in data analysis [16].

Bottom Line: Here we propose a new analysis method, PhenoDissim, which computes the phenotypic dissimilarity between cell populations via Support Vector Machine classification and cross validation.PhenoDissim is a novel analysis method for image-based high-throughput screen, relying on two parameters which can be automatically optimized without a priori knowledge.PhenoDissim is freely available as an R package.

View Article: PubMed Central - HTML - PubMed

Affiliation: German Cancer Research Center (DKFZ), Div, Signaling and Functional Genomics and Department of Cell and Molecular Biology, Medical Faculty Mannheim, Im Neuenheimer Feld 580, D-69120 Heidelberg, Germany. xianzhang@gmail.com.

ABSTRACT

Background: Discovering functional relationships of genes through cell-based phenotyping has become an important approach in functional genomics. High-throughput imaging offers the ability to quantitatively assess complex phenotypes after perturbation by RNA interference (RNAi). Such image-based high-throughput RNAi screening studies have facilitated the discovery of novel components of gene networks and their interactions. Images generated by automated microscopy are typically analyzed by extracting quantitative features of individual cells, resulting in large multidimensional data sets. Robust and sensitive methods to interpret these data sets and to derive biologically relevant information in a high-throughput and unbiased manner remain to be developed.

Results: Here we propose a new analysis method, PhenoDissim, which computes the phenotypic dissimilarity between cell populations via Support Vector Machine classification and cross validation. Applying this method to a kinome RNAi screening data set, we demonstrate that the proposed method shows a good replicate reproducibility, separation of controls and clustering quality, and we are able to identify siRNA phenotypes and discover potential functional links between genes.

Conclusions: PhenoDissim is a novel analysis method for image-based high-throughput screen, relying on two parameters which can be automatically optimized without a priori knowledge. PhenoDissim is freely available as an R package.

Show MeSH
Related in: MedlinePlus