Multi-Parameter Optimization: Explore How We Made an Already Exceptional Protein Better


Abdulaziz Elgammal

November 27, 2022


Integrating Ankh and other proprietary models, our Protein engineering platform (PEP) is capable of multi-parameter protein optimization of complex proteins. Here, we showcase how we have integrated several components of our platform to engineer a protein integral in many biotechnology industries to enhance multiple parameters, stability, and solubility

Blog content:

  1. Overview of the Project 
  2. Hits Generation
  3. Lead Selection & Validation
  4. Outperforming the Wildtype and Commercial Alternatives

Overview of the Project:

Multi-parameter protein optimization is a complex engineering problem. It involves extensive finetuning of protein amino acids towards an optimum point in the protein function landscape. Going through a protein optimization problem, a couple of problems start to emerge.

Optimizing one function of the protein is already an extensive and expensive task. A typical protein has a couple of hundreds of amino acids with 19 different possible changes in each location. In this complex problem, you are left with a significant number of variations to try. 

The standard toolkit has two options:

  • Random mutagenesis: which is typically random shots on the same goal

The approach is extensively utilized but has significant drawbacks, including the time and cost needed to generate and test each variant physically.

  • Rational design: analysis of the structure to narrow down some guiding rules

The approach requires plenty of skills and a significant time commitment to analyze residues individually.

Proteinea’s Protein Engineering Platform (PEP) capitalizes on the advantages of both platforms being comprehensive and knowledge-based simultaneously. Our process involves training our AI models for a better understanding of protein sequence to structure to function relationships to achieve multiparameter protein optimization in significantly less time and less cost.

From generating meaningful hits to lead selection and validation, we can speed up the process significantly while having a higher probability of success.

Hits Generation: 

Ankh - Meaningful Protein Representation: To model the language of life, we need a meaningful representation encompassing its structural and functional information. We used our top-performing state-of-art protein language model Ankh to represent our target.

Hidden Gems Database Exclusive Access to Proteins and their Meaning: To build our growth factor training dataset, we leveraged Hidden Gems, our massive database with exclusive access to proteins, holistic representations, and annotations.

Gem MiningConquering Sparse & Novel Targets: To expand our training dataset, we utilized Gem Mining, our original search algorithm landing hits with fundamental structural and functional similarity with often low sequence similarity!

AnkhGen – In Silico Directed Evolution: To generate top-performing hits, We employed our patented approach that expands our protein goldmines from size n to size n2 to navigate hundreds of thousands of possible mutations for function-guided generation.

Lead Selection & Validation:

To deliver high-confidence variants for experimental validation, We exploited our holistic filtration engine to look for leads via a multi-layered and multi-path sequence and structural analysis. We were able to filter through hundreds of thousands of variants for the top performing leads based on the desired parameters.

For experimental testing we integrated our cloud and external labs to reinforce our computational promotion of leads outperforming top commercially available variants. Our rapid high-throughput experimental validation through our Conan PI screening platform is essential to enhancing our computational performance and empowering further iterations. We input our smart experimental feedback through our computational models to achieve even better results in subsequent

Outperforming the Wildtype and Commercial Alternatives:

Focusing on the protein thermal stability, here we show that thermal stability was enhanced itrativly over two iterations with a significant improvement in protein’s thermal stability. Additionally, the protein yield increased to over 4X (not shown) while increase or maintaining the biological functionality of the protein. 

Iterative cycles of engineering takes the exceptional results into new hights.

Let’s explore how we can make your exceptional protein better

Contact us at for more information on commercial partnerships

Latest blogs

Ankh: Pioneering the paradigm shifting technology of protein language models

To model the language of life (i.e., the language of proteins) in a data-efficient and cost-optimized modality, we need a meaningful representation encompassing its structural and functional information. Proteinea has developed -and is continuously improving- Ankh, a revolutionary base protein language model that will change the landscape of protein engineering with AI forever

Image of a big dna molecule, closeup

The Power of Codon Optimization: Lower immunogenicity and aggregation with exceptionally higher yield

Proteinea codon optimization managed to increase the yield of antibody expression 300 - 600% while increasing protein stability and reducing immunogenicity. This's all due to our codon optimization models CodonSP, no single amino acid change.

View All

Our innovations