A feature selection approach for identification of signature genes from SAGE data
2007

Identifying Signature Genes from SAGE Data

Sample size: 24 publication Evidence: moderate

Author Information

Author(s): Barrera Junior, Cesar Roberto M Jr, Humes Carlos Jr, Martins David C Jr, Patrão Diogo FC, Silva Paulo JS, Brentani Helena

Primary Institution: Instituto de Matemática e Estatística, Universidade de São Paulo

Hypothesis

Can a new framework effectively select specific genes that distinguish different biological states using SAGE data?

Conclusion

The proposed methodology effectively identifies signature genes that can separate different biological states using SAGE data.

Supporting Evidence

  • The methodology allows for the identification of genes that can distinguish between different tumor grades.
  • The study highlights the importance of using credibility intervals for more robust gene selection.
  • Results showed that the proposed method can be adapted for other counting methods like MPSS and SBS.

Takeaway

This study created a new way to find important genes that help tell different types of brain tumors apart using a special data analysis method.

Methodology

The study used a new framework that applies bolstered error estimation and credibility intervals to select genes from SAGE data.

Limitations

The small sample size and high data variability make it difficult to define the 'best triple' of genes.

Participant Demographics

The study included 24 SAGE libraries from different tumor types: 2 from normal brain, 4 from astrocytoma grade II, 9 from astrocytoma grade III, and 9 from glioblastoma.

Digital Object Identifier (DOI)

10.1186/1471-2105-8-169

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication