Abstract
A common approach for understanding the relationship between transcription factors (TFs) and transcription factor binding sites (TFBSs) is to use features at either the TF level or the DNA level. For a given TF family, features can be derived from the DNA-binding domains at the protein level as well as TF binding sites at the DNA sequence level. Here we investigate the relative importance of features from these different levels for main TF families to better understand: (1) family-specific features and (2) the proportion of features from either the DNA or protein level. We perform class-wise feature selection on TF families to identify important features for each family. Importance of the selected features is assessed in terms of predictive accuracy of assigning TFs and associated TFBSs to correct TF families. Evaluation of the best model on an independent test set resulted in a predictive accuracy of ∼90%. Analysis of the selected features used in the best model on a family-by-family basis shows congruence with the fact that interaction between TF proteins and TFBS in the DNA is quite family specific. Our analysis further suggests that: (1) this approach can be used to determine and better understand which features (at both the DNA and protein levels) are important to consider for each TF family, and (2) a similar approach to combine DNA and protein level features may be useful for other datasets where protein-DNA interaction is a key component of biological function.
Original language | English (US) |
---|---|
Pages (from-to) | 2097-2102 |
Number of pages | 6 |
Journal | Pattern Recognition Letters |
Volume | 31 |
Issue number | 14 |
DOIs | |
State | Published - Oct 15 2010 |
Externally published | Yes |
Bibliographical note
Funding Information:The authors acknowledge financial support offered by the Agency for Science, Technology, and Research, Singapore (A∗Star) under Grant #052 101 0020 .
Keywords
- Feature selection
- Multi-class classification
- TF family-specific features
- TF-TFBS interaction
- TFBS
- Transcription factor
ASJC Scopus subject areas
- Software
- Signal Processing
- Computer Vision and Pattern Recognition
- Artificial Intelligence