Max Valentín Staller

This paper develops interpretable biophysical neural networks to predict transcriptional activation domains from protein sequences, revealing how residue properties influence both activation strength and protein abundance.

The paper shows that function can be conserved in transcriptional activation domains without sequence conservation through evolutionary turnover of key residues.

This study shows that short activation domains, not just DNA-binding domains, significantly influence chromatin association of transcription factors in living cells.

The paper develops ADhunter, a model to identify and quantify transcriptional activators in fungal genomes, expanding functional annotations across non-model species.

The paper investigates how structural ensembles of disordered activation domains correlate with transcriptional activity through mutational analysis of HIF-1 and CITED2.

The paper identifies and validates activation domains in non-transcription factor proteins across Arabidopsis and yeast, discovering novel coactivators and universal genetic parts for synthetic biology.

The paper predicts acidic transcriptional activation domains from protein sequences by analyzing the balance of acidic and hydrophobic residues, showing improved accuracy when combining human and yeast-based models.

The paper identifies three design principles of acidic transcriptional activation domains through experimental analysis and develops a predictor to identify these domains in the human proteome.

The paper shows that Caudal counter-represses Hunchback to enable even-skipped stripe 2 expression in Drosophila embryos, challenging the traditional view of Hunchback as a direct activator.

This study uses a deep mutational scan to analyze the sequence features of an acidic activation domain, finding that acidic residues help expose hydrophobic motifs and that activity increases under amino acid starvation.