The 2020s Will be to Biotech What the 1970s Were to Computing
DECEMBER 1ST, 2020
Yesterday Google DeepMind announced that their deep learning system AlphaFold has achieved unprecedented levels of accuracy on the “protein folding problem”, a grand challenge problem in computational biochemistry.
The ability to accurately predict protein structures from their amino-acid sequence represents a huge boon to life sciences and medicine. It potentially vastly accelerates efforts to understand the building blocks of cells and enable quicker and more advanced drug discovery.
For decades, laboratory experiments have been the main way to get good protein structures. The first complete structures of proteins were determined, starting in the 1950s, using a technique in which X-ray beams are fired at crystallized proteins and the diffracted light translated into a protein’s atomic coordinates. X-ray crystallography has produced the lion’s share of protein structures.
AlphaFold’s breakthrough could mean that lower-quality and easier-to-collect experimental data, in combination with AlphaFold models, would be all that’s needed to get a good structure, essentially democratizing access to highly accurate protein structures.
“AlphaFold is a once in a generation advance, predicting protein structures with incredible speed and precision. This leap forward demonstrates how computational methods are poised to transform research in biology and hold much promise for accelerating the drug discovery process.”
Above are two examples of protein targets in the free modelling category. AlphaFold predicts highly accurate structures measured against experimental result. A protein’s shape is closely linked with its function, and the ability to predict this structure unlocks a greater understanding of what it does and how it works. Many of the world’s greatest challenges, like developing treatments for diseases or manufacturing enzymes that break down industrial waste, are fundamentally tied to protein design.
Two main applications concerning protein folding are recombiant protein engineering and the de novo design of proteins with novel functions. Engineered recombinant proteins that are used in the clinic today include recombinant hormones, interferons, interleukins, growth factors, tumor necrosis factors, blood clotting factors, thrombolytic drugs, and enzymes for treating major diseases such as diabetes, dwarfism, myocardial infarction, congestive heart failure, cerebral apoplexy, multiple sclerosis, neutropenia, thrombocytopenia, anemia, hepatitis, rheumatoid arthritis, asthma, Crohn’s disease and cancers therapies.
Therapeutic recombinant proteins have undergone multiple generations, with noticeable improvement in each generation. The first generation of recombinant proteins included proteins with native structures, while the second generation involved proteins with improved properties, especially PK, biodistribution, specificity, efficacy, and minimized side effects. The third generation, aided by a significantly better ability to model protein folding, will include recombinant proteins that have been edited for novel routes of administration, include new formulations, exhibit higher efficiency and increased safety.
The de novo design of proteins with novel functions carries with it the hope of constructing proteins with functions unprecedented in nature. The lynchpins of this research are 1.) our understanding of the principles of protein folding 2.) our ability to edit genetic sequences (CRISPR) to engineer new proteins. To date, with a limited toolset, de novo proteinmaking has yielded an experimental HIV vaccine, novel proteins that aim to combat all strains of the influenza viruses simultaneously, carrier molecules that can ferry reprogrammed DNA into cells, and new enzymes that help microbes suck carbon dioxide out of the atmosphere and convert it into useful chemicals. Recently, researchers announced the sucessful synthesis of de novo proteins that spontaneously arrange themselves in a flat layer, like interlocking tiles on a bathroom floor. Such surfaces may lead to novel types of solar cells and electronic devices. Nobody fully knows the implications of de novo protein design because it has the potential to impact dozens of different disciplines.
Attempting to predict the impact of widespread de novo protein synthesis would be akin to attempting to predict how widespread computer usage would transform our world in the 1970s. Similarly, the impact of widespread de novo protein engineering has the potential to be revolutionary across almost every sector of the economy. The efforts of this field were accelerated significantly yesterday by the team at Google AlphaFold.
