As we push through the fall semester, take a minute to learn about some of the recent discoveries and developments in drug discovery, quantum computing and cancer treatment.
Deep-learning method accelerates drug discovery
An Oct. 23 paper in Science detailed the development of a deep-learning model, DrugReflector, as a computational drug discovery tool. Traditionally, drug discovery has been a tedious process of manually screening compounds on the bench, but DrugReflector acts as a preliminary screening step to computationally predict how specific compounds may affect a cell’s activity and gene expression. The authors of the study reported 17x increase in compound hit rate compared to manual screening methods when they used DrugReflector to screen for drugs affecting blood cell production.
Since DrugReflector is capable of learning, researchers were able to feed the results of their downstream drug testing step back to the model, which further improved its ability to predict novel drug targets. However, it’s important to note that the program was initially trained on only around 9,600 compounds across 50 cell types, which is far smaller than the number of chemical compounds available for drug use.
Google announces a new powerful quantum algorithm
Researchers at Google have announced that they have reached “quantum advantage” with their Willow quantum chip in an Oct. 22 press release. Quantum advantage, or quantum supremacy, is the ability for a quantum computer to solve a problem inaccessible by classical computers. The basis of quantum computing is the use of units known as qubits, which are quantum particles able to exist in more states than traditional computing’s binary bits.
Google researchers claim that their algorithm, known as Quantum Echoes, is able to find the structure of basic molecules by simulating the interaction between the molecule’s nuclear magnetic spins 13,000x faster than traditional computers. Though Quantum Echoes seems promising, researchers not involved in the project are wary to accept Google’s result without more extensive analysis — it is still to be determined if Google’s quantum algorithm is truly superior to traditional algorithms. Furthermore, there is still a considerable amount of work to be done before quantum computing can be used for useful applications including drug discovery, agriculture, materials science, mathematics and physics.
COVID-19 vaccines may lead to cancer resistance
A study in Nature showed that non-small cell lung cancer (NSCLC) and melanoma patients who received a dose of an mRNA-based COVID-19 vaccine had higher cancer survival rates. The researchers conducted mouse experiments and found that the mRNA COVID-19 vaccine increased the activity of the immune system and sensitized tumors to immune checkpoint inhibitors, a common immunotherapy used to treat a wide range of cancers. In humans, receiving the vaccine within 100 days of starting checkpoint inhibitor therapy was associated with significantly increased survival rates compared to patients who either did not receive the vaccine or received it outside of the 100-day range.
The authors proposed that the mRNA vaccines used produced a more potent immune response because the mRNA was directly transported into cells instead of only into the bloodstream, making them better able to prime T cells to attack tumors. This preliminary research could lead to the further development of immune-stimulating mRNA-based vaccines effective against a broad range of cancers.