Prediction and functional characterization of transcriptional activation domains

Mahatma et al. leveraged a neural network approach to systematically predict transcription factor activation domains. The findings further highlight the importance of aromatic and negatively charged amino acids and reveal the importance of unknown amino acid properties.