# AI Model Predicts Protein Shapes from Infrared Light Data > This patent describes a computer method that uses an artificial intelligence model to predict the detailed 3D shapes of proteins within a food ingredient by analyzing how the ingredient absorbs infrared light. - **Patent:** US 10962473 - **Original title:** Protein secondary structure prediction - **Owner:** Notco Delaware - **Granted:** 2021 - **Status:** Active - **Times cited:** 14 - **Field:** biotech, food_science, software, ai_ml, materials ## What it does The patent outlines a computer-based method for predicting the secondary structure of a specific protein in an ingredient. First, it collects two types of digital data: Fourier Transform Infrared Spectroscopy (FTIR) spectra from many ingredients and their known protein secondary structures (Claim 1). The FTIR data is then transformed into a 'quantized' format. This prepared data trains an artificial intelligence (AI) model. Once trained, the AI model receives the FTIR spectrum of a new, specific ingredient and predicts its protein secondary structure. For example, a food company could use this to quickly determine if a new plant-based protein ingredient has the desired alpha-helix or beta-sheet structures for a specific texture or function in a product. ## What it does NOT cover - Does not cover predicting protein secondary structure without using FTIR spectra as input data. - Does not cover methods that predict protein secondary structure without using an artificial intelligence model. - Does not cover AI models that do not include multiple sub-models like Partial Least Squares or XGBoost, as specified in Claim 3. - Does not cover predicting protein tertiary or quaternary structures, only secondary structures like alpha-helix and beta-sheet. - Does not cover methods that do not involve a training phase using a dataset of known FTIR spectra and corresponding secondary structures. ## The clever bit The novelty lies in combining specific machine learning techniques, including multiple sub-models like XGBoost and Lasso regression within a 'stacking ensemble regressor,' to accurately interpret complex FTIR spectra and predict protein secondary structures. This approach allows for a more robust and precise prediction than simpler models. ## Real-world examples 1. Quality control in food manufacturing for protein-rich ingredients 2. Development of new plant-based protein products 3. Characterization of protein ingredients for nutritional supplements 4. Research into protein functionality in biopharmaceuticals ## Why it matters Understanding protein secondary structure is crucial in fields like food science, pharmaceuticals, and materials science, as it directly influences a protein's function, stability, and interaction with other molecules. This patent offers a rapid, non-destructive way to gain this insight, potentially speeding up research and development cycles. For instance, in developing plant-based foods, knowing the protein structure helps engineers formulate products with specific textures, like the chewiness of meat alternatives. ## Frequently asked questions ### What does AI Model Predicts Protein Shapes from Infrared Light Data cover? This patent describes a computer method that uses an artificial intelligence model to predict the detailed 3D shapes of proteins within a food ingredient by analyzing how the ingredient absorbs infrared light. ### Who owns patent US 10962473? Notco Delaware owns this patent, granted in 2021. ### When does this patent expire? This patent is expected to expire on November 5, 2040, when the invention enters the public domain. ### What is patent US 10962473 cited by? This patent has been cited by 14 later patents that build on its ideas. ### What problem does this patent solve? Understanding protein secondary structure is crucial in fields like food science, pharmaceuticals, and materials science, as it directly influences a protein's function, stability, and interaction with other molecules. This patent offers a rapid, non-destructive way to gain this insight, potentially speeding up research and development cycles. For instance, in developing plant-based foods, knowing the protein structure helps engineers formulate products with specific textures, like the chewiness of meat alternatives. ### What does this patent NOT cover? Does not cover predicting protein secondary structure without using FTIR spectra as input data. **Full plain-English explainer:** https://patentbrief.org/patent/us/10962473/protein-secondary-structure-prediction **Original patent:** https://patents.google.com/patent/US10962473 --- _Source: PatentBrief — https://patentbrief.org. Patent facts are from public records; the plain-English explanation is PatentBrief's._ ## Related patents Semantically similar inventions in the PatentBrief corpus: - [AI Predicts 3D Protein Shapes from Genetic Code](https://patentbrief.org/patent/us/20240412809/protein-structure-prediction-from-amino-acid-sequences-using-self-attention-neur) — This patent describes how a computer uses self-attention neural networks to predict the complex three-dimensional shape of a protein based on its simple amino acid sequence, a process crucial for understanding biology and developing new medicines. - [How to Predict Protein Shapes Better Using Lab Tests and Computers](https://patentbrief.org/patent/us/20210174903/enhanced-protein-structure-prediction-using-protein-homolog-discovery-and-constr) — This patent describes a method to improve predicting a protein's 3D shape by combining computer simulations with actual distance measurements from specific parts of the protein in a lab. - [Using AI to Predict Protein Shapes for Drug Discovery](https://patentbrief.org/patent/us/12437843/predicting-protein-structures-using-geometry-neural-networks-that-estimate-simil) — This patent describes an AI-driven method for iteratively predicting protein 3D structures using a geometry neural network, then using those predictions to find and synthesize new drug molecules. - [How Computers Predict Protein Shapes Faster Using Smart Templates](https://patentbrief.org/patent/us/20210280268/protein-structure-prediction-system) — This patent describes a computer method to quickly predict the 3D shape of proteins by creating and refining "synthetic templates" from existing protein structures, reducing the heavy computational work usually needed. - [Computer Model Designs Proteins by Matching Structure to Sequence](https://patentbrief.org/patent/us/12412637/embedding-based-generative-model-for-protein-design) — IBM's 2025 patent describes a computer system that designs new protein sequences by first understanding a target 3D protein shape and then generating a sequence that would fold into that shape.