Differential gene expression patterns in cells of the mammalian brain result in the morphological,connectional, and functional diversity of cells. A wide variety of studies have shown that certaingenes are expressed only in specific cell-types.
Analysis of cell-type-specific gene expressionpatterns can provide insights into the relationship between genes, connectivity, brain regions, andcell-types. However, automated methods for identifying cell-type-specific genes are lacking to date.
Results:
Here, we describe a set of computational methods for identifying cell-type-specific genes in themouse brain by automated image computing of in situ hybridization (ISH) expression patterns.
Weapplied invariant image feature descriptors to capture local gene expression information fromcellular-resolution ISH images. We then built image-level representations by applying vectorquantization on the image descriptors.
We employed regularized learning methods for classifyinggenes specifically expressed in different brain cell-types. These methods can also rank imagefeatures based on their discriminative power.
We used a data set of 2,872 genes from the Allen BrainAtlas in the experiments. Results showed that our methods are predictive of cell-type-specificity ofgenes.
Our classifiers achieved AUC values of approximately 87% when the enrichment level is setto 20. In addition, we showed that the highly-ranked image features captured the relationshipbetween cell-types.
Conclusions:
Overall, our results showed that automated image computing methods could potentially be used toidentify cell-type-specific genes in the mouse brain.
Author: Rongjian LiWenlu ZhangShuiwang Ji
Credits/Source: BMC Bioinformatics 2014, 15:209
Published on: 2014-06-20
Tweet
News Provider: 7thSpace Interactive
Social Bookmarking
RETWEET This! | Digg this! | Post to del.icio.us | Post to Furl | Add to Netscape | Add to Yahoo! | Rojo
There are no comments available. Be the first to write a comment.