When it comes to AI, can we Ditch the Datasets?

MIT researchers have demonstrated the use of a generative machine-learning model to create synthetic data, based on real data, that can be used to train another model for image classification. This image shows examples of the generative model’s transformation methods. Credit: Massachusetts Institute of Technology

Huge amounts of data are needed to train machine-learning models to perform image classification tasks, such as identifying damage in satellite photos following a natural disaster. However, these data are not always easy to come by. Datasets may cost millions of dollars to generate, if usable data exist in the first place, and even the best datasets often contain biases that negatively impact a model’s performance.

To circumvent some of the problems presented by datasets, M...

Read More

Combing the Cosmos: New Color Catalog Aids Hunt for Life on Frozen Worlds

With a color catalog based on Earth’s microbes, astronomers can begin to decipher the tint of life on distant, frozen exoplanets, as depicted in this artistic rendering by Jack Madden Ph.D. ’20.

Aided by microbes found in the subarctic conditions of Canada’s Hudson Bay, an international team of scientists has created the first color catalog of icy planet surface signatures to uncover the existence of life in the cosmos.

As ground-based and space telescopes get larger and can probe the atmosphere of rocky exoplanets, astronomers need a color-coded guide to compare them and their moons to vibrant, tinted biological microbes on Earth, which may dominate frozen worlds that circle different stars.

But researchers need to know what microbes that live in frigid places on Earth look ...

Read More

Spider Silk can Stabilize Cancer-Suppressing Protein

•p53 is poorly expressed and conformationally unstable
•A spider silk domain boosts p53 translation in vitro
•The spider silk-p53 fusion protein adopts a compact state and is biologically active
•Reducing N-terminal disorder in fusion proteins increases expression and stability

The p53 protein protects our cells from cancer and is an interesting target for cancer treatments. The problem is, however, that it breaks down rapidly in the cell. Researchers at Karolinska Institutet in Sweden have now found an unusual way of stabilising the protein and making it more potent. By adding a spider silk protein to p53, they show that it is possible to create a protein that is more stable and capable of killing cancer cells. The study is published in the journal Structure.

P53 plays a key r...

Read More

The Next Generation of Robots will be Shape-Shifters

Active matter: Wrapping an elastic ball (orange) in a layer of tiny robots (blue) allows researchers to program shape and behaviour. Image credit: Jack Binysh

Physicists have discovered a new way to coat soft robots in materials that allow them to move and function in a more purposeful way. The research, led by the UK’s University of Bath, is described today in Science Advances.

Authors of the study believe their breakthrough modelling on ‘active matter’ could mark a turning point in the design of robots. With further development of the concept, it may be possible to determine the shape, movement and behaviour of a soft solid not by its natural elasticity but by human-controlled activity on its surface.

The surface of an ordinary soft material always shrinks into a sphere...

Read More