Patents by Inventor Walid Moustafa

Walid Moustafa has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • AA2CDS:pre-trained amino acid-to-codon sequence mapping enabling efficient expression and yield optimization
    Patent number: 12380966
    Abstract: Systems, methods, and apparatus for generating a DNA/gene sequence output, given a protein sequence input. In one aspect, the protein-to-DNA mapping system includes a protein language model (embedder) that generates a high-dimensional protein embedding from a low-dimensional protein sequence input; and the system includes a protein-to-DNA translator that processes the protein embedding and generates a DNA sequence output, based on the protein embedding. The protein-to-DNA translator is a neural network (e.g., a Seq2Seq model) and is configured with translation coefficients that may be trained with a training dataset, comprising the training protein and DNA embeddings. Training the translation coefficients can include supervised learning that maps the protein embedding (input) to a DNA sequence (target).
    Type: Grant
    Filed: July 10, 2024
    Date of Patent: August 5, 2025
    Assignee: Proteinea, Inc.
    Inventors: Hazem Essam, Hosna Eltarras, Wafaa Ashraf Salah-Eldin, Mohamed Soudy, Ahmed Saleh, Walid Moustafa, Ahmed Eid Zoheir, Mohamed Ashraf Elkerdawy
  • GeneCull: Enabling High-Quality Gene Sequence Modeling via Evolution-Guided Data Pruning Criteria
    Publication number: 20250046397
    Abstract: A computer-implemented method is provided. The computer implemented method includes generating high-yield training datasets by excluding those genetic sequences that do not satisfy at least one abundance condition, at least one stability condition, at least one expression condition, and/or at least one translation efficiency condition. The computer-implemented method further includes training at least one model on the high-yield training datasets to generate high-yield genetic sequences.
    Type: Application
    Filed: August 3, 2024
    Publication date: February 6, 2025
    Applicant: Proteinea, Inc.
    Inventors: Wafaa Ashraf SALAH-ELDIN, Mohamed Ashraf ELKERDAWY, Mohamed SOUDY, Ahmed EID ZOHEIR, Mohamed ELSHARKAWY, Hosna ELTARRAS, Walid Moustafa, Abdulaziz ELGAMMAL
  • AA2CDS:Pre-trained Amino Acid-to-Codon Sequence Mapping Enabling Efficient Expression and Yield Optimization
    Publication number: 20250022540
    Abstract: Systems, methods, and apparatus for generating a DNA/gene sequence output, given a protein sequence input. In one aspect, the protein-to-DNA mapping system includes a protein language model (embedder) that generates a high-dimensional protein embedding from a low-dimensional protein sequence input; and the system includes a protein-to-DNA translator that processes the protein embedding and generates a DNA sequence output, based on the protein embedding. The protein-to-DNA translator is a neural network (e.g., a Seq2Seq model) and is configured with translation coefficients that may be trained with a training dataset, comprising the training protein and DNA embeddings. Training the translation coefficients can include supervised learning that maps the protein embedding (input) to a DNA sequence (target).
    Type: Application
    Filed: July 10, 2024
    Publication date: January 16, 2025
    Applicant: Proteinea, Inc.
    Inventors: Hazem ESSAM, Hosna ELTARRAS, Wafaa Ashraf SALAH-ELDIN, Mohamed SOUDY, Ahmed SALEH, Walid Moustafa, Ahmed EID ZOHEIR, Mohamed Ashraf ELKERDAWY