Identifying spatially similar gene expression patterns in early stage fruit fly embryo images: binary feature versus invariant moment digital representations

Abstract

Background: Modern developmental biology relies heavily on the analysis of embryonic gene expression patterns. Investigators manually inspect hundreds or thousands of expression patterns to identify those that are spatially similar and to ultimately infer potential gene interactions. However, the rapid accumulation of gene expression pattern data over the last two decades, facilitated by high-throughput techniques, has produced a need for the development of efficient approaches for direct comparison of images, rather than their textual descriptions, to identify spatially similar expression patterns.

Comparison of embryonic expression within multigene families using the flyexpress discovery platform reveals more spatial than temporal divergence

“…In this case, the similarity score ( S ) between two images ( Q and D ) is given by S QD = | Q ∩ D |/| Q ∪ D |, where | Q ∩ D | is the size of the intersection of expression (count of black pixels) between images Q and D and | Q ∪ D| is the size of the union between images Q and D . We have previously shown that this approach emphasizes spatial overlap, which is biologically more meaningful than shape matching and invariant moment based features (Kumar et al, 2002; Gurunathan et al, 2004). We have also found that it performs with an effectiveness similar to the computationally more intensive Gaussian Mixture Model method (Peng et al, 2007), which in our hands is very sensitive to shifts in image properties, such as the color and contrast (Gargesha et al, 2008; Gargesha et al, 2009a; Gargesha et al, 2009b; Roy et al, 2009).…”

“…A common first step in discovering gene interactions is to identify genes with overlapping expression patterns. However, the standard practice of manual inspection of images is not efficient given the extraordinary number of images available today (Gurunathan et al, 2004; Peng et al, 2007; Walter et al, 2010). This problem has been addressed utilizing textual descriptions of gene expression images employing controlled vocabularies (CVs; Janning, 1997; Brody, 1999; FlyBase, 1999; Drysdale, 2001; Tomancak et al, 2002; Matthews et al, 2005; Grumbling and Strelets, 2006).…”

“…Further, no textual annotations exist for a plethora of spatial expression patterns due to the tremendous resources required, as assigning such annotations must be done manually (Gurunathan et al, 2004). Furthermore, under many circumstances it is not feasible to fully describe an expression pattern in words alone.…”

“…Sequence data can be easily expressed in four or twenty building blocks (for DNA and proteins, respectively), but images containing spatial expression patterns show remarkable variation in orientation, size, lighting conditions, and color-spectrum depending on the laboratory techniques used. Thus, biological image analysis is more challenging and a new frontier of computational biology (Gurunathan et al, 2004; Peng et al, 2007; Walter et al, 2010). …”

FlyExpress 7: An Integrated Discovery Platform To Study Coexpressed Genes Usingin SituHybridization Images inDrosophila

“…Black-and-white spatial profiles (representing gene expression in the developing embryo) were created for each image. The black-and-white spatial profiles translate into Binary Feature Vectors representing the composite spatial expression of each gene or genomic element ( Gurunathan et al , 2004 ). Images were annotated by genotype, image source, Campos-Ortega stage ( Campos-Ortega and Hartenstein, 1985 ), and anatomical orientation.…”