Drug development encompasses multiple processes, wherein protein subcellular localization is essential. It promotes target identification, treatment development, and the design of drug delivery systems. In this research, a deep learning framework called LocPro is presented for predicting protein subcellular localization. Specifically, LocPro is unique in (a) combining protein representations from the pre-trained large language model (LLM) ESM2 and the expert-driven tool PROFEAT, (b) implementing a hybrid deep neural network architecture that integrates convolutional neural network (CNN), fully connected (FC) layer, and bidirectional long short-term memory (BiLSTM) blocks, and (c) developing a multi-label framework for predicting protein subcellular localization at multiple granularity levels. Additionally, a dataset was curated and divided using a homology-based strategy for training and validation. Comparative analyses show that LocPro outperforms existing methods in sequence-based multi-label protein subcellular localization prediction. The practical utility of this framework is further demonstrated through case studies on drug target subcellular localization. All in all, LocPro serves as a valuable complement to existing protein localization prediction tools. The web server is freely accessible at https://idrblab.org/LocPro/.

Objective: To investigate the relationship between tropomodulin 3(TMOD3) and the malignant biological characteristics of hepatocellular carcinoma, and the predictive potential of TMOD3 as a biomarker for the recurrence and metastasis of hepatocellular carcinoma. Methods: Firstly, the structure of TMOD3 and its subcellular localization in cells and tissues were analyzed using database of Human Protein Atlas. Then explored the differential expression of TMOD3 in hepatocellular carcinoma tissues and normal liver tissues and its impact on clinical pathological characteristics and prognosis using TCGA and GEO datasets. Subsequently, the STRING database was utilized to explore the interacting proteins of TMOD3, followed by enrichment analysis conducted using the Metascape database. Finally, Logistic regression was used to analyze the independent risk factors for metastasis of hepatocellular carcinoma and evaluated the importance of predictive variables using ROC curves and Wald tests. Survival analysis was conducted using the Kaplan-Meier curve and the Log-rank test. Results: TMOD3 was localized to actin filaments in cells, and compared with normal tissues, the expression level of TMOD3 in liver cancer tissues is higher(P<0.05), and high expression of TMOD3 was closely related to lymph node metastasis and distant metastasis of hepatocellular carcinoma patients(P<0.05). Enrichment analysis results revealed that TMOD3 and its interacting proteins mainly function and signaling pathways related to tumor invasion and migration. Logistic regression found that TMOD3 was an independent risk factor for recurrence and metastasis of hepatocellular carcinoma(OR: 4.359, 95%CI: 1.235-15.384,P=0.022). Survival analysis revealed that high expression of TMOD3 was associated with poor OS, DFS, and RFS of hepatocellular carcinoma patients(P<0.05). Both ROC analysis and Wald test indicated that TMOD3 has good predictive characteristics for recurrence and metastasis of hepatocellular carcinoma. Conclusion TMOD3 is closely associated with the invasion and metastasis of hepatocellular carcinoma and is an independent risk factor for the recurrence and metastasis of liver cancer. TMOD3 performs well in predicting the recurrence and metastasis of hepatocellular carcinoma and has the potential to become a biomarker for predicting the recurrence and metastasis of hepatocellular carcinoma.

Malignant tumors have become one of the major diseases that seriously endanger human life.Early diagnosis and treatment can greatly improve the survival rate of patients.Imaging examinations based on fluorescence imaging,CT,photoacoustic imaging,MRI,and PET have been widely studied and applied in the diagnosis of tumors.However,early cancerous tissue and normal tissue have similar imaging signals,which is difficult to be accurately distinguished by conventional imaging.With the development and cross integration of physics,materials science,biology,and medicine,nanomaterials have shown broad application prospects in the diagnosis and treatment of diseases due to their unique physical and chemical properties.The enhanced permeability and retention effect in solid tumor,and the easy modification properties of nanomaterials allow them to accurately"recognize"tumor and accelerate its enrichment at the tumor site;the imaging characteristics allow them to be used as contrast agents to enhance the signal intensity of the tumor site;their responsiveness mechanism can also allow them to distinguish normal from cancerous cells according to the microenvironment in tumor cells.In addition,multimodality imaging based on nanomaterials can compensate for the shortcomings of single modality imaging and achieve real-time and omnidirectional imaging of tumors.With multiple functions integrated,nanomaterials are expected to enhance the imaging signals of early cancerous sites,improve signal-to-noise ratio,and achieve early diagnosis of tumors.