• Natural precursor utilization

• Alternative synthesis methods

• Atom economy optimization

• Toxicity assessment

• Comparative lifecycle analysis (benchmarking)

• Real-time monitoring

• Scalability design

• Phytochemistry

• Sustainable design

The rise of organic electronics still faced many challenges to compete with inorganic counterparts. To advance the progress, computational simulation and modelling can help give insight to working principles and its design principle. However, structural diversity of molecular configuration such as orientation or distances can make traditional modelling difficult. It is due to the fact that intermolecular and intramolecular factors can critically influence electronic charge-transfer coupling. Although machine learning can be one of the solution, existing ML models still struggle with accuracy for complex structural variation and often trained with crystalline systems. This research aim to solve the problem using two-step machine learning; combining a model to predicts molecular orbitals of individual molecule fragments and second model to predict electronic coupling from MO overlap integrals.

The G protein-coupled receptor (GPCR) superfamily is one of the largest and most diverse groups of proteins in the human genome. It plays a crucial role in cellular signaling and response to external stimuli. Unfortunately, existing databases like UniProt or GPCRdb still lack panoramic and interactive views of GPCR sequence-structure-function relationships. Moreover, due to the complex relationship within the GPCR family, it is difficult to classify new GPCR sequences and understand their roles within the superfamily network.  Here, we provide a web-based graphical database: SeQuery; with multiple resolutions (individual, sequences, clusters, and families). The tools provide integrated GPCR sequences, structures, and functional annotations from various databases. Network analysis is calculated using BLASTp for similarity and distance, and minimum span clustering (MSC) to obtain cluster hierarchy at three resolution levels. Interactive visualization built with Cytoscape.js and centrality metrics used to identify key nodes in the network.

Point-of-care detection can be important in an emergency or when an outbreak occurs. However, most high-end detection facilities require complex and relatively large devices with additional sample preparation that limits access to remote places. Using the surface plasmon resonance (SPR) principle, the research aims to build a portable, easy-to-use biosensor. Real-time detection can be achieved thanks to the output signal that can be interpreted directly with high detection sensitivity.