Predicting housekeeping genes based on Fourier analysis.
basic_science · Level V
Where this comes from
- Record sourced from PubMed, PMID 21687628.
- Also identified by DOI 10.1371/journal.pone.0021012 and PMC identifier PMC3133835.
- No licence information is recorded for this record.
- Because redistribution is not established, this page shows the abstract only. Follow the links below for the full text.
Abstract
Housekeeping genes (HKGs) generally have fundamental functions in basic biochemical processes in organisms, and usually have relatively steady expression levels across various tissues. They play an important role in the normalization of microarray technology. Using Fourier analysis we transformed gene expression time-series from a Hela cell cycle gene expression dataset into Fourier spectra, and designed an effective computational method for discriminating between HKGs and non-HKGs using the support vector machine (SVM) supervised learning algorithm which can extract significant features of the spectra, providing a basis for identifying specific gene expression patterns. Using our method we identified 510 human HKGs, and then validated them by comparison with two independent sets of tissue expression profiles. Results showed that our predicted HKG set is more reliable than three previously identified sets of HKGs.