Dataset Information

Mining expressed sequence tags identifies cancer markers of clinical interest.

ABSTRACT:

Background

Gene expression data are a rich source of information about the transcriptional dis-regulation of genes in cancer. Genes that display differential regulation in cancer are a subtype of cancer biomarkers.

Results

We present an approach to mine expressed sequence tags to discover cancer biomarkers. A false discovery rate analysis suggests that the approach generates less than 22% false discoveries when applied to combined human and mouse whole genome screens. With this approach, we identify the 200 genes most consistently differentially expressed in cancer (called HM200) and proceed to characterize these genes. When used for prediction in a variety of cancer classification tasks (in 24 independent cancer microarray datasets, 59 classifications total), we show that HM200 and the shorter gene list HM100 are very competitive cancer biomarker sets. Indeed, when compared to 13 published cancer marker gene lists, HM200 achieves the best or second best classification performance in 79% of the classifications considered.

Conclusion

These results indicate the existence of at least one general cancer marker set whose predictive value spans several tumor types and classification types. Our comparison with other marker gene lists shows that HM200 markers are mostly novel cancer markers. We also identify the previously published Pomeroy-400 list as another general cancer marker set. Strikingly, Pomeroy-400 has 27 genes in common with HM200. Our data suggest that a core set of genes are responsive to the deregulation of pathways involved in tumorigenesis in a variety of tumor types and that these genes could serve as transcriptional cancer markers in applications of clinical interest. Finally, our study suggests new strategies to select and evaluate cancer biomarkers in microarray studies.

SUBMITTER: Campagne F

PROVIDER: S-EPMC1635568 | biostudies-literature | 2006 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Mining expressed sequence tags identifies cancer markers of clinical interest.

Campagne Fabien F Skrabanek Lucy L

BMC bioinformatics 20061101

<h4>Background</h4>Gene expression data are a rich source of information about the transcriptional dis-regulation of genes in cancer. Genes that display differential regulation in cancer are a subtype of cancer biomarkers.<h4>Results</h4>We present an approach to mine expressed sequence tags to discover cancer biomarkers. A false discovery rate analysis suggests that the approach generates less than 22% false discoveries when applied to combined human and mouse whole genome screens. With this ap ...[more]

PMID: 17078886

Similar Datasets

Project description:Simple Sequence Repeats (SSRs) are widely used in population genetic studies but their classical development is costly and time-consuming. The ever-increasing available DNA datasets generated by high-throughput techniques offer an inexpensive alternative for SSRs discovery. Expressed Sequence Tags (ESTs) have been widely used as SSR source for plants of economic relevance but their application to non-model species is still modest.Here, we explored the use of publicly available ESTs (GenBank at the National Center for Biotechnology Information-NCBI) for SSRs development in non-model plants, focusing on genera listed by the International Union for the Conservation of Nature (IUCN). We also search two model genera with fully annotated genomes for EST-SSRs, Arabidopsis and Oryza, and used them as controls for genome distribution analyses. Overall, we downloaded 16 031 555 sequences for 258 plant genera which were mined for SSRsand their primers with the help of QDD1. Genome distribution analyses in Oryza and Arabidopsis were done by blasting the sequences with SSR against the Oryza sativa and Arabidopsis thaliana reference genomes implemented in the Basal Local Alignment Tool (BLAST) of the NCBI website. Finally, we performed an empirical test to determine the performance of our EST-SSRs in a few individuals from four species of two eudicot genera, Trifolium and Centaurea.We explored a total of 14 498 726 EST sequences from the dbEST database (NCBI) in 257 plant genera from the IUCN Red List. We identify a very large number (17 102) of ready-to-test EST-SSRs in most plant genera (193) at no cost. Overall, dinucleotide and trinucleotide repeats were the prevalent types but the abundance of the various types of repeat differed between taxonomic groups. Control genomes revealed that trinucleotide repeats were mostly located in coding regions while dinucleotide repeats were largely associated with untranslated regions. Our results from the empirical test revealed considerable amplification success and transferability between congenerics.The present work represents the first large-scale study developing SSRs by utilizing publicly accessible EST databases in threatened plants. Here we provide a very large number of ready-to-test EST-SSR (17 102) for 193 genera. The cross-species transferability suggests that the number of possible target species would be large. Since trinucleotide repeats are abundant and mainly linked to exons they might be useful in evolutionary and conservation studies. Altogether, our study highly supports the use of EST databases as an extremely affordable and fast alternative for SSR developing in threatened plants.

Project description:BackgroundPearl millet [Pennisetum glaucum (L.) R. Br.] is a staple food and fodder crop of marginal agricultural lands of sub-Saharan Africa and the Indian subcontinent. It is also a summer forage crop in the southern USA, Australia and Latin America, and is the preferred mulch in Brazilian no-till soybean production systems. Use of molecular marker technology for pearl millet genetic improvement has been limited. Progress is hampered by insufficient numbers of PCR-compatible co-dominant markers that can be used readily in applied breeding programmes. Therefore, we sought to develop additional SSR markers for the pearl millet research community.ResultsA set of new pearl millet SSR markers were developed using available sequence information from 3520 expressed sequence tags (ESTs). After clustering, unigene sequences (2175 singlets and 317 contigs) were searched for the presence of SSRs. We detected 164 sequences containing SSRs (at least 14 bases in length), with a density of one per 1.75 kb of EST sequence. Di-nucleotide repeats were the most abundant followed by tri-nucleotide repeats. Ninety primer pairs were designed and tested for their ability to detect polymorphism across a panel of 11 pairs of pearl millet mapping population parental lines. Clear amplification products were obtained for 58 primer pairs. Of these, 15 were monomorphic across the panel. A subset of 21 polymorphic EST-SSRs and 6 recently developed genomic SSR markers were mapped using existing mapping populations. Linkage map positions of these EST-SSR were compared by homology search with mapped rice genomic sequences on the basis of pearl millet-rice synteny. Most new EST-SSR markers mapped to distal regions of linkage groups, often to previous gaps in these linkage maps. These new EST-SSRs are now are used by ICRISAT in pearl millet diversity assessment and marker-aided breeding programs.ConclusionThis study has demonstrated the potential of EST-derived SSR primer pairs in pearl millet. As reported for other crops, EST-derived SSRs provide a cost-saving marker development option in pearl millet. Resources developed in this study have added a sizeable number of useful SSRs to the existing repertoire of circa 100 genomic SSRs that were previously available to pearl millet researchers.

Project description:BACKGROUND: With the advent of low-cost, high-throughput sequencing, the amount of public domain Expressed Sequence Tag (EST) sequence data available for both model and non-model organism is growing exponentially. While these data are widely used for characterizing various genomes, they also present a serious challenge for data quality control and validation due to their inherent deficiencies, particularly for species without genome sequences. DESCRIPTION: ConiferEST is an integrated system for data reprocessing, visualization and mining of conifer ESTs. In its current release, Build 1.0, it houses 172,229 loblolly pine EST sequence reads, which were obtained from reprocessing raw DNA sequencer traces using our software--WebTraceMiner. The trace files were downloaded from NCBI Trace Archive. ConiferEST provides biologists unique, easy-to-use data visualization and mining tools for a variety of putative sequence features including cloning vector segments, adapter sequences, restriction endonuclease recognition sites, polyA and polyT runs, and their corresponding Phred quality values. Based on these putative features, verified sequence features such as 3' and/or 5' termini of cDNA inserts in either sense or non-sense strand have been identified in-silico. Interestingly, only 30.03% of the designated 3' ESTs were found to have an authenticated 5' terminus in the non-sense strand (i.e., polyT tails), while fewer than 5.34% of the designated 5' ESTs had a verified 5' terminus in the sense strand. Such previously ignored features provide valuable insight for data quality control and validation of error-prone ESTs, as well as the ability to identify novel functional motifs embedded in large EST datasets. We found that "double-termini adapters" were effective indicators of potential EST chimeras. For all sequences with in-silico verified termini/terminus, we used InterProScan to assign protein domain signatures, results of which are available for in-depth exploration using our biologist-friendly web interfaces. CONCLUSION: ConiferEST represents a unique and complementary public resource for EST data integration and mining in conifers by reprocessing raw DNA traces, identifying putative sequence features and determining and annotating in-silico verified features. Seamlessly integrated with other public resources, ConiferEST provides biologists powerful tools to verify data, visualize abnormalities, including EST chimeras, and explore large EST datasets.

Dataset Information

Mining expressed sequence tags identifies cancer markers of clinical interest.

Background

Results

Conclusion

Publications

Mining expressed sequence tags identifies cancer markers of clinical interest.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets