Nico Callewaert, Morgane Boone, and their group from VIB and Ghent University devised a new technique to analyze the secretability of hundreds of thousands of protein sequences at once.
Morgane Boone. Image Credit: ©UCSF.
This novel tool facilitates the training of (deep) machine learning predictors of protein secretability and enables unparalleled research works on eukaryotic protein secretion. Using this new technique in applied settings can enhance the predictability of protein manufacturing in biotechnology.
Biopharma and biotechnology rely on recombinant protein production, ideally in a form produced from the host cells, to simplify the purification mechanism. But obtaining detectable levels of functional recombinant protein produced by a heterologous host is yet a process of trial and error.
Measuring protein secretion in high throughput
The specific features that allow or stop the production of proteins were unclear. Recently, researchers headed by Nico Callewaert and Morgane Boone (VIB-UGent Center for Medical Biotechnology) created a novel tool, SECRiFY, which was designed to start bringing much predictability to the field. Researchers, for the first time, have a technique that can be used to generate yeast secretability data of protein fragments on a proteome-wide scale.
Machine learning models are trained on these data to augur the secretability of proteins. This predictability is based on the primary sequence and higher-order features derived from the sequence, or patterns, learned from the sequence.
The current research combines novel protein molecular biological method development in the wet laboratory with advanced protein sequence/structure machine learning in the laboratories of Lennart Martens and Sven Degroeve (VIB-UGent), Wim Vranken (VUB), and Wesley De Neve (Ghent University Global Campus, Incheon, South-Korea).
Easing production of recombinant proteins by predicting their secretion
The SECRiFY tool offers an understanding of the sequence properties that enables secretory processing of short protein sequences. This allows new fundamental research into eukaryotic protein secretion.
This novel technique and its continuing expansion to bigger protein domains and proteins of direct biotechnological interest can offer insights into the features that influence secretability. It also helps comprehend the rules protein sequences must follow for successful passage through the yeast secretory system.
Also, bigger databases are being created on proteomes of biomedical interest, full of the experimental evidence on which protein chunks can be produced from biotech’s favorite yeast systems. Eventually, this understanding can specifically boost the experimental expression and manufacturability of proteins of biotechnological interest.
With the current advent of technologies for cost-effective on-demand synthesis of large, customized libraries of coding sequences, SECRiFY analysis of those is bound to enrich our models of protein secretability rapidly.”
Nico Callewaert, Center for Medical Biotechnology, Vlaams Instituut voor Biotechnologie
“For example, in the burgeoning field of computational protein design, adding the dimension of biotechnological manufacturability of such protein variants should speed up progress to practical applications in biomedicine, biocatalysts for the greening of the economy, and many more,” adds Nico Callewaert.
Boone, M., et al. (2021) Massively parallel interrogation of protein fragment secretability using SECRiFY reveals features influencing secretory system transit. Nature Communications. doi.org/10.1038/s41467-021-26720-y.