Artificial intelligence (AI) is reshaping diverse fields in science, with molecular science being no exception. A recent study published in Research (Science Partner Journal) reports S²ALM, a powerful AI tool which uses a step-wise learning approach for a detailed analysis of antibodies. Using both 1D sequence and 3D structural data for antibody learning, the developed model outperforms prior models across key tasks, advancing antibody design for various diseases and redefining the future of therapeutic development.

To address the growing plastic crisis, researchers have developed a catalytic method that breaks down common plastic waste—polyethylene, polypropylene, and their mixtures—into valuable olefins used in fuel and chemical production. The process uses inexpensive base-metal catalysts and achieves up to 95% carbon recovery at just 320 °C—significantly milder than conventional thermal recycling. This energy-efficient strategy offers a promising alternative to traditional methods, which are energy intensive or rely on expensive noble-metal catalysts.

Biological tissues like skin, arteries, and cartilage have a non-linear strain-stiffening relationship. Some biomimetic hydrogel scaffolds have been successful in effectively replicating this behavior. However, achieving structural complexity in such strain-stiffening hydrogels has been difficult. A recent Research study has demonstrated an innovative and efficient technique, immersion phase separation 3D printing, to fabricate structurally complex tissues with strain-stiffening properties. These hydrogel scaffolds can pave the way for biomimetic, patient-specific implants in the future.

Current robotic grippers employ soft and flexible materials to mimic human like grasping behavior. However, they require continuous energy input to maintain their grasp, limiting practical applications. In a new study, researchers develop an innovative bio-inspired bistable robotic gripper, that maintains its grasp with no energy input. It can also adjust the force required for switching between open and closed states, making it suitable for diverse tasks.

Multimodal sentiment analysis is an information processing technique that attempts to predict human emotional states from multiple modalities like text, audio, and video. Due to challenges in aligning multiple modalities, existing methods are limited to analysis at course or fine granularity, which risks missing nuances in human emotional expression. Researchers have now developed an innovative approach to MSA that reduces computational time required to sentiment prediction while offering improved performance.