The objective of the project is to develop innovative algorithms based on deep learning (machine learning) techniques to use available sequence and functional data to predict optimized biologics in silico. The basis for this method will be developed by using peptides as a model case but will be extended to more complex proteins like antibodies in a second phase. The final product is a user-friendly software, which allows researchers the in silico optimization of their biologics under development in house.
Users are pharmaceutical or biotech companies as well as academic research groups active in the development of novel biologics (therapeutics or diagnostics).
The company and two other German partners (universities) have the expertise in the development of algorithms and software, deep learning methods, NGS analysis and chip-based high-throughput binding analysis of initial candidates.
The partner SME, which is sought to complement the consortium, should have programs for the development of peptides (pure or enclosed in a scaffold), provide enriched libraries for NGS, and perform secondary functional tests of selected peptides in its own systems. The partner would greatly benefit by enhancing its own research programs and having access to our bioinformatic know-how.
Details for EuroTransBio 12th transnational call that applies here:
The partner SME should be located in Austria, Flanders (Belgium), Finland, France (Alsace, Champagne-Ardenne and Lorraine), Italy, or Russia due to the restrictions of the current EuroTransBio 12th transnational call.
The deadline for EOIs is 9th of December 2016 (although earlier EOIs are highly encouraged because of the winter holidays).
The deadline for the submission of the proposal is January 31st, 2017.
Anticipated start of the project is in the middle of 2017.
A partner is sought who brings an actual development project (diagnostic or therapeutic) into the cooperation since they will not only profit in their development efforts, but the novel product is an additional innovation within the grant project. This should be beneficial for enhancing the chances of success in the EuroTransBio call.
The partner should develop the peptides as described under technical specifications by selection methods like phage, yeast, or bacterial display and provide enriched libraries for next-generation sequencing analysis and the target molecule for binding analysis.
The SME partner that is sought will do the bioinformatic analysis and identify enriched sequence families. A German partner (university) will test the candidates in a chip-based high-throughput format for binding kinetics.
These data are used to develop algorithms by means of machine learning (deep learning) to suggest optimized sequence candidates, which should exhibit better binding kinetics but are as well optimized for other properties, e.g. effective production (2nd German partner university).
The partner, who is sought here, should further test the candidates in his functional assays and provide the data for further optimization of the algorithms. The consortium leader will do the final development of the software.