The field of molecular modeling and drug discovery is rapidly advancing, with a focus on developing innovative methods for molecular property prediction, protein design, and ligand generation. Recent research has highlighted the importance of integrating structural and sequential information to improve the accuracy and generalizability of molecular models. Diffusion-based models have shown promise in generating realistic and diverse protein structures and ligands, while graph neural networks and transformer architectures have been used to improve the prediction of molecular properties and protein-ligand interactions. Noteworthy papers in this area include SigmaDock, which introduces a novel fragmentation scheme and SE(3) Riemannian diffusion model for molecular docking, and SPECTRA, which presents a spectral target-aware graph augmentation framework for imbalanced molecular property regression. Other notable papers include Peptide2Mol, DGTN, and SiDGen, which demonstrate the potential of deep generative models and diffusion-based methods for molecular design and protein-ligand interaction prediction.