EST expression profiling provides an attractive tool for studying differential gene expression, but cDNA libraries' origins and EST data quality are not always known or reported. Libraries may originate from pooled or mixed tissues; EST clustering, EST counts, library annotations and analysis algorithms may contain errors. Traditional data analysis methods, including research into tissue-specific gene expression, assume EST counts to be correct and libraries to be correctly annotated, which is not always the case. Therefore, a method capable of assessing the quality of expression data based on that data alone would be invaluable for assessing the quality of EST data and determining their suitability for mRNA expression analysis. Here we report an approach to the selection of a small generic subset of 244 UniGene clusters suitable for identification of the tissue of origin for EST libraries and quality control of the expression data using EST expression information alone. We created a small expression matrix of UniGene IDs using two rounds of selection followed by two rounds of optimisation. Our selection procedures differ from traditional approaches to finding “tissue-specific” genes and our matrix yields consistency high positive correlation values for libraries with confirmed tissues of origin and can be applied for tissue typing and quality control of libraries as small as just a few hundred total ESTs. Furthermore, we can pick up tissue correlations between related tissues e.g. brain and peripheral nervous tissue, heart and muscle tissues and identify tissue origins for a few libraries of uncharacterised tissue identity. It was possible to confirm tissue identity for some libraries which have been derived from cancer tissues or have been normalised. Tissue matching is affected strongly by cancer progression or library normalisation and our approach may potentially be applied for elucidating the stage of normalisation in normalised libraries or for cancer staging.

BackgroundThe Cancer Genome Anatomy Project (CGAP) xProfiler and cDNA Digital Gene Expression Displayer (DGED) have been made available to the scientific community over a decade ago and since then were used widely to find genes which are differentially expressed between cancer and normal tissues. The tissue types are usually chosen according to the ontology hierarchy developed by NCBI. The xProfiler uses an internally available flat file database to determine the presence or absence of genes in the chosen libraries, while cDNA DGED uses the publicly available UniGene Expression and Gene relational databases to count the sequences found for each gene in the presented libraries.ResultsWe discovered that the CGAP approach often includes libraries from dependent or irrelevant tissues (one third of libraries were incorrect on average, with some tissue searches no correct libraries being selected at all). We also discovered that the CGAP approach reported genes from outside the selected libraries and may omit genes found within the libraries. Other errors include the incorrect estimation of the significance values and inaccurate settings for the library size cut-off values. We advocated a revised approach to finding libraries associated with tissues. In doing so, libraries from dependent or irrelevant tissues do not get included in the final library pool. We also revised the method for determining the presence or absence of a gene by searching the UniGene relational database, revised calculation of statistical significance and sorted the library cut-off filter.ConclusionOur results justify re-evaluation of all previously reported results where NCBI CGAP expression data and tools were used.

No abstract

No abstract

There are currently a few bioinformatics tools, such as dbEST, DDD, GEPIS, cDNA xProfiler and cDNA DGED to name a few, which have been widely used to retrieve and analyse EST expression data and for comparing gene expression levels e.g. between cancer and normal tissues. The outcome of any such comparison depends on EST libraries' annotations and assumes that the actual expression data (EST counts) are correct. None of the existing tools provide a quality control method for the selection and evaluation of the original EST expression libraries. Here we report the selection, optimisation and evaluation of a minimal gene expression data set using CGAP cDNA DGED. Our approach relies solely on the expression data itself and is independent on the libraries annotations. The reported approach allows tissue typing of expression libraries of different sizes containing between as little as 249 total EST counts and up to 13,929 total EST counts (the highest tested so far). A NEW APPROACH TO THE QUALITY CONTROL OF EXPRESSION DATATissue phenotype depends on gene expression as 278 Milnthorpe A. and Soloviev M..Optimisation and Validation of a Minimum Data Set for the Identification and Quality Control of EST Expression Libraries.