Arabidopsis thaliana, commonly known as thale cress, is a widely studied model organism in plant biology. Understanding the subcellular localization of proteins within Arabidopsis cells is crucial for unraveling their functions and deciphering complex cellular processes. Subcellular localization provides valuable insights into protein trafficking, interactions, and regulatory mechanisms, ultimately aiding in the identification of potential targets for agricultural and biotechnological applications. Advancements in high-throughput sequencing technologies have led to an exponential increase in available genomic and proteomic data. However, experimental determination of protein localization remains labor-intensive and time-consuming. Consequently, computational methods that can accurately predict protein subcellular localization have gained significant attention. AtSubP-2.0 deep neural network for Arabidopsis thaliana subcellular localization prediction. The AtSubP-2.0 model utilizes a multi-layered architecture with a combination of convolutional neural networks (CNNs) and recurrent neural networks (RNNs). It takes protein sequence inputs and generates predictions for localization to various cellular compartments, including the nucleus, cytoplasm, plasma membrane, endoplasmic reticulum, Golgi apparatus, mitochondria, peroxisomes, and vacuoles. Accurate subcellular localization prediction is crucial for unraveling protein functions and biological processes in plants. The AtSubP-2.0 deep neural network provides a powerful tool for the Arabidopsis research community, aiding in the efficient annotation and characterization of protein localization, thereby accelerating our understanding of plant cellular biology.