Last week, the Nobel Committees announced the much-anticipated Nobel Prize winners of 2024, honouring the contributions that, as per Alfred Nobel's will of 1895, "have conferred the greatest benefit to humankind".
In this series of articles, we present selected patents of these winners that, at least to some extent, result from or lead to their celebrated works. For those with an interest in the most commemorated scientific and technological achievements in 2024 and intellectual property, this series makes for interesting reading.
On 9 October 2024, the Royal Swedish Academy of Sciences announced that the 2024 Nobel Prize in chemistry was rewarded to David Baker for "for computational protein design", jointly with Demis Hassabis and John M. Jumper "for protein structure prediction".
The Committee commented that:
Proteins generally consist of 20 different amino acids, which can be described as life's building blocks. In 2003, David Baker succeeded in using these blocks to design a new protein that was unlike any other protein. Since then, his research group has produced one imaginative protein creation after another, including proteins that can be used as pharmaceuticals, vaccines, nanomaterials and tiny sensors.
...
In 2020, Demis Hassabis and John Jumper presented an AI model called AlphaFold2. With its help, they have been able to predict the structure of virtually all the 200 million proteins that researchers have identified. Since their breakthrough, AlphaFold2 has been used by more than two million people from 190 countries. Among a myriad of scientific applications, researchers can now better understand antibiotic resistance and create images of enzymes that can decompose plastic.
These discoveries have enabled protein structures prediction and design, which significantly benefits humankind.
Below we highlight two patent families with Baker and Jumper listed as co-inventors.
David Baker | Self-assembling protein nanostructures displaying paramyxovirus and/or pneumovirus F proteins and their use
This patent family claims priority to US Provisional Application No. 62/481,331 and has an earliest priority date of 4 April 2017. PCT Application No. PCT/US2018/025880 includes nine independent claims.
Claims are defined below:
- A nanostructure, comprising:
a plurality of first assemblies, each first assembly comprising a plurality of identical first poly peptides
a plurality of second assemblies, each second assembly comprising a plurality of identical second polypeptides, wherein the second polypeptide differs from the first polypeptide
wherein the plurality of first assemblies non-covalently interact with the plurality of second assemblies to form a nanostructure, and
wherein the nanostructure displays multiple copies of one or more paramyxovirus and/or pneumovirus F proteins, or antigenic fragments thereof, on an exterior of the nanostructure.
- A method for generating an immune response to paramyxovirus and/or pneumovirus F protein in a subject, comprising administering to the subject in need thereof an effective amount of the nanostructure or immunogenic composition of any one of claims 1-29 and 34-35 to generate the immune response.
- A method for treating or limiting a paramyxovirus and/or pneumovirus infection in a subject, comprising administering to the subject in need thereof an effective amount of the nanostructure or immunogenic composition of any one of claims 1-29 and 34-35 to, thereby treating or preventing paramyxovirus and/or pneumovirus infection in the subject.
The other independent claims include those directed to a recombinant nucleic acid, a recombinant expression vector, a recombinant host cell, an immunogenic composition and a process for assembling the nanostructures.
The invention relates to synthetic nanostructures and methods of designing such nanostructures. The first polypeptides and the second polypeptides are non-naturally occurring proteins that can be produced by any suitable means, including recombinant production or chemical synthesis.
There are no specific primary amino acid sequence requirements for the first and second polypeptides. The nanostructures can be used for generating an immune response to paramyxovirus and/or pneumovirus F protein in a subject, and/or treating or limiting a paramyxovirus and/or pneumovirus infection.
This patent family entered national phase in various jurisdictions including Australia, Brazil, Canada, China, Europe, Israel, Japan, Korea, Mexico, The Philippines, Russia, Singapore and the United States.
John M. Jumper | Machine learning for determining protein structures
This patent family claims priority from US Provisional Applications No. 62/734,757, 62/734,773 and 62/770, 490 and has an earliest priority date of 21 September 2018. PCT Application No. PCT/EP2019/074670 has seven independent claims.
Claims are defined below:
- A method performed by one or more data processing apparatus
for determining a final predicted structure of a given protein,
wherein the given protein includes a sequence of amino acids,
wherein a predicted structure of the given protein is defined by
values of a plurality of structure parameters, the method
comprising:
obtaining initial values of the plurality of structure parameters defining the predicted structure;
updating the initial values of the plurality of structure parameters, comprising, at each of a plurality of update iterations:
determining a quality score characterizing a quality of the predicted structure defined by current values of the structure parameters, wherein the quality score is based on respective outputs of one or more scoring neural networks which are each configured to process: (i) the current values of the structure parameters, (ii) a representation of the sequence of amino acids of the given protein, or (iii) both; and
for one or more of the plurality of structure parameters:
determining a gradient of the quality score with respect to the current value of the structure parameter; and
updating the current value of the structure parameter using the gradient of the quality score with respect to the current value of the structure parameter; and
determining the predicted structure of the given protein to be defined by the current values of the plurality of structure parameters after a final update iteration of the plurality of update iterations; and
selecting a particular predicted structure of the given protein as the final predicted structure of the given protein.
- A method performed by one or more data processing apparatus
for determining a predicted structure of a given protein, wherein
the given protein includes a sequence of amino acids, wherein the
predicted structure of the given protein is defined by values of a
plurality of structure parameters, the method
comprising:
obtaining initial values of the plurality of structure parameters defining the predicted structure;
updating the initial values of the plurality of structure parameters, comprising, at each of a plurality of update iterations:
determining a quality score characterizing a quality of the predicted structure defined by current values of the structure parameters, wherein the quality score is based on respective outputs of one or more scoring neural networks which are each configured to process: (i) the current values of the structure parameters, (ii) a representation of the sequence of amino acids of the given protein, or (iii) both;
for one or more of the plurality of structure parameters:
determining a gradient of the quality score with respect to the current value of the structure parameter; and
updating the current value of the structure parameter using the gradient of the quality score with respect to the current value of the structure parameter;
determining the predicted structure of the given protein to be defined by the current values of the plurality of structure parameters after a final update iteration of the plurality of update iterations.
The other independent claims include those directed to a method of obtaining a ligand, a method of obtaining a polypeptide ligand, a method of identifying the presence of a protein mis-folding disease and relevant computer storage media storage devices and storing instructions.
Recognising that the biological function of a protein is determined by its structure and determining protein structures, may facilitate understanding life processes and the design of proteins. This invention relates to a system and a method of predicting protein structures.
For example, the method and system involve processing data to define an amino acid sequence of a certain protein, using machine learning algorithms to generate a final predicted structure of the protein. The final predicted structure defines an estimate of a three-dimensional configuration of the atoms in the amino acid sequence of the protein after the protein undergoes protein folding.
The invention vastly improves on previous methods of determining protein structures using physical experiments, which can be time-consuming and expensive. The invention may be used in drug development, as the protein structure can be used to determine how drugs bind to a protein.
This patent family entered national phase in various jurisdictions including Canada, China, Europe, Japan and the United States.
Hassabis does not appear to be a co-inventor of any patent applications directly in relation to protein structure prediction.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
