Modeling Protein-DNA Interactions with Feature Motif Models
Author Information
Author(s): Sharon Eilon, Lubliner Shai, Segal Eran
Primary Institution: Weizmann Institute of Science
Hypothesis
Can feature motif models (FMMs) better represent transcription factor (TF) binding specificities than position specific scoring matrices (PSSMs)?
Conclusion
Feature motif models explain TF binding significantly better than traditional PSSMs.
Supporting Evidence
- FMMs were shown to explain TF binding data better than PSSMs on both synthetic and real data.
- The study developed a motif finder algorithm that learns FMM motifs from unaligned sequences.
- FMMs capture dependencies between positions that PSSMs cannot.
Takeaway
This study shows a new way to understand how proteins stick to DNA, which helps us learn more about how genes are controlled.
Methodology
The study developed a probabilistic method for modeling TF-DNA interactions using feature motif models that capture dependencies between binding positions.
Potential Biases
Potential biases may arise from the datasets used, which could affect the generalizability of the findings.
Limitations
The study primarily focuses on specific datasets and may not generalize to all TFs or conditions.
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website