SCANN Project

[Stem Cell characterisation by Artificial Neural Network]

Ghislain Bidaut ^1,2,*
Chris Stoeckert ¹

1: Center for Bioinformatics, Department of Genetics, University of Pennsylvania School of Medicine, Philadelphia, PA 19104, USA
2: Integrative Bioinformatics, Centre de Recherche en Cancérologie de Marseille, INSERM U891 - Institut Paoli-Calmettes, 27 Boulevard Leï Roure, 13009 Marseille, France

Introduction

It is believed that stem cells differentiation and fate are triggered by a common molecular program that gives those cells the ability to differentiate into specialized progenitors and fully differentiated cells. To extract the stemness signature of several cells types at the transcriptional level, we integrated heterogeneous microarray experiments performed in different adult and embryonic tissues (liver, blood, bone, prostate and stomach in Homo sapiens and Mus musculus). Data were integrated by generalization of the hematopoietic stem cell hierarchy and by homology between mouse and human. The variation-filtered and integrated gene expression dataset was fed to a single-layered neural network to create a classifier to (i) extract the stemness signature and (ii) characterize unknown stem cell tissue samples by attribution of a stem cell differentiation stage. We were able to characterize mouse stomach progenitor and human prostate progenitor samples and isolate gene signatures playing a fundamental role for every level of the stem cell generalized hierarchy. Cross-validation training led to a 63 gene signature minimizing classsification error on the input dataset.

Publications

Bidaut G, Stoeckert CJ Jr. (in prep.) Large Scale Transcriptome Data Integration Across Multiple Tissues to Decipher Stem Cell Signatures. Methods in Enzymology.
Bidaut G, Stoeckert CJ Jr. Characterization of unknown adult stem cell samples by large scale data integration and artificial neural networks. Pac Symp Biocomput. 2009:356-67.

Material

Input dataset is avalable here [ *.txt.gz | *.zip ]

The list of markers for every category is available here [ weight-matrix-63-genes.txt.gz | weight-matrix-63-genes.zip ].

The full data are (input dataset and list of markers) is avalable as a single package here. [scann-data.tar.gz].

The complete SCGAP database is available here with downloads to individual datasets.

Source Code

Source code (version 1.0) of the classifier can be downloaded here [scann-1.0.tar.gz] - a machine capable of interpreting perl code is needed in order to run it.

Questions? Comments ? Please do not hesitate to contact directly the corresponding author (Ghislain Bidaut) from the sourceforge web site.