Dataset Information

Extraction and comparison of gene expression patterns from 2D RNA in situ hybridization images.

ABSTRACT:

Motivation

Recent advancements in high-throughput imaging have created new large datasets with tens of thousands of gene expression images. Methods for capturing these spatial and/or temporal expression patterns include in situ hybridization or fluorescent reporter constructs or tags, and results are still frequently assessed by subjective qualitative comparisons. In order to deal with available large datasets, fully automated analysis methods must be developed to properly normalize and model spatial expression patterns.

Results

We have developed image segmentation and registration methods to identify and extract spatial gene expression patterns from RNA in situ hybridization experiments of Drosophila embryos. These methods allow us to normalize and extract expression information for 78,621 images from 3724 genes across six time stages. The similarity between gene expression patterns is computed using four scoring metrics: mean squared error, Haar wavelet distance, mutual information and spatial mutual information (SMI). We additionally propose a strategy to calculate the significance of the similarity between two expression images, by generating surrogate datasets with similar spatial expression patterns using a Monte Carlo swap sampler. On data from an early development time stage, we show that SMI provides the most biologically relevant metric of comparison, and that our significance testing generalizes metrics to achieve similar performance. We exemplify the application of spatial metrics on the well-known Drosophila segmentation network.

Availability

A Java webstart application to register and compare patterns, as well as all source code, are available from: http://tools.genome.duke.edu/generegulation/image_analysis/insitu

Contact

uwe.ohler@duke.edu

Supplementary information

Supplementary data are available at Bioinformatics online.

SUBMITTER: Mace DL

PROVIDER: S-EPMC3140183 | biostudies-literature | 2010 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Extraction and comparison of gene expression patterns from 2D RNA in situ hybridization images.

Mace Daniel L DL Varnado Nicole N Zhang Weiping W Frise Erwin E Ohler Uwe U

Bioinformatics (Oxford, England) 20091126 6

<h4>Motivation</h4>Recent advancements in high-throughput imaging have created new large datasets with tens of thousands of gene expression images. Methods for capturing these spatial and/or temporal expression patterns include in situ hybridization or fluorescent reporter constructs or tags, and results are still frequently assessed by subjective qualitative comparisons. In order to deal with available large datasets, fully automated analysis methods must be developed to properly normalize and ...[more]

PMID: 19942587

Similar Datasets

Project description:Spontaneously occurring canine mammary gland tumors share many features with human breast cancer, including biological behavior and histologic features. Compared to transgenic murine model, canine models have advantages including naturally occurring models of human diseases and cancer. In humans, breast cancer is divided into molecular subtypes based on ER, PR, and HER2 expression. In contrast with humans, few studies have evaluated these subtypes in canine mammary gland tumors, including expression of HER2. HER2 expression in canine mammary tissues has been further complicated by controversy regarding the antibody's specificity. This study aimed to investigate c-erbB2 mRNA expression in retrospective formalin-fixed paraffin embedded samples, using RNA in situ hybridization with a novel quantitative assay and to compare this method with immunohistochemistry. Using 48 canine mammary tumor samples and 14 non-neoplastic canine mammary tissues, RNA in situ hybridization was performed with RNAscope® using a canine-specific target gene probe (ERBB2), and quantitative measurement was performed using the housekeeping gene (POLR2A) to calculate the target gene/housekeeping gene ratio. The ratio of ERBB2/POLR2A was quantified using open-source image analysis programs and compared with the immunohistochemistry results. A significant correlation was observed between the HER2 immunohistochemistry score and ERBB2/POLR2A RNA in situ hybridization (P < 0.001). When the HER2 immunohistochemistry score was 3+, significantly higher expression of HER2 mRNA was observed by RNA in situ hybridization. Interestingly, HER2 mRNA was also observed in non-neoplastic mammary tissues by RNA in situ hybridization. This assay potentially facilitates the reliable quantification of mRNA expression levels in retrospective formalin-fixed paraffin-embedded samples. Further studies are required to elucidate the role of HER2 in canine mammary gland tumors and to implement clinical trials in dogs.

Project description:Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease that causes death within a mean of 2-3 years from symptom onset. There is no diagnostic test and the delay from symptom onset to diagnosis averages 12 months. The identification of prognostic and diagnostic biomarkers in ALS would facilitate earlier diagnosis and faster monitoring of treatments. Gene expression profiling (GEP) can help to identify these markers as well as therapeutic targets in neurological diseases. One source of genetic material for GEP in ALS is peripheral blood, which is routinely accessed from patients. However, a high proportion of globin mRNA in blood can mask important genetic information. A number of methods allow safe collection, storage and transport of blood as well as RNA stabilisation, including the PAXGENE and TEMPUS systems for the collection of whole blood and LEUKOLOCK which enriches for the leukocyte population. Here we compared these three systems and assess their suitability for GEP in ALS. We collected blood from 8 sporadic ALS patients and 7 controls. PAXGENE and TEMPUS RNA extracted samples additionally underwent globin depletion using GlobinClear. RNA was amplified and hybridised onto Affymetrix U133 Plus 2.0 arrays. Lists of genes differentially regulated in ALS patients and controls were created for each method using the R package PUMA, and RT-PCR validation was carried out on selected genes. TEMPUS/GlobinClear, and LEUKOLOCK produced high quality RNA with sufficient yield, and consistent array expression profiles. PAXGENE/GlobinClear yield and quality were lower. Globin depletion for PAXGENE and TEMPUS uncovered the presence of over 60% more transcripts than when samples were not depleted. TEMPUS/GlobinClear and LEUKOLOCK gene lists respectively contained 3619 and 3047 genes differentially expressed between patients and controls. Real-time PCR validation revealed similar reliability between these two methods and gene ontology analyses revealed similar pathways differentially regulated in disease compared to controls.

Dataset Information

Extraction and comparison of gene expression patterns from 2D RNA in situ hybridization images.

Motivation

Results

Availability

Contact

Supplementary information

Publications

Extraction and comparison of gene expression patterns from 2D RNA in situ hybridization images.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure