DeepSP: A Deep Learning Framework for Spatial Proteomics

Abstract

The study of protein subcellular localization (PSL) is a fundamental step toward understanding the mechanism of protein function. The recent development of mass spectrometry (MS)-based spatial proteomics to quantify the distribution of proteins across subcellular fractions provides us a high-throughput approach to predict unknown PSLs based on known PSLs. However, the accuracy of PSL annotations in spatial proteomics is limited by the performance of existing PSL predictors based on traditional machine learning algorithms. In this study, we present a novel deep learning framework named DeepSP for PSL prediction of an MS-based spatial proteomics data set. DeepSP constructs the new feature map of a difference matrix by capturing detailed changes between different subcellular fractions of protein occupancy profiles and uses the convolutional block attention module to improve the prediction performance of PSL. DeepSP achieved significant improvement in accuracy and robustness for PSL prediction in independent test sets and unknown PSL prediction compared to current state-of-the-art machine learning predictors. As an efficient and robust framework for PSL prediction, DeepSP is expected to facilitate spatial proteomics studies and contributes to the elucidation of protein functions and the regulation of biological processes.

TransGCN: a semi-supervised graph convolution network–based framework to infer protein translocations in spatio-temporal proteomics

“…A common challenge in model training is that it tends to lead to under-learning of a few classes by the model when dealing with large differences in the amount of data from different classes [ 56 ]. In this study, we investigated the effects of sample imbalance in the model classification problem by introducing a sample weight factor to optimize the model [ 12 ]. By adjusting the sample weight factor, we obtained improved prediction accuracy and improved generalization performance.…”

“…The spatio-temporal proteomics includes at least one paired control and treated experiments. It is often necessary to collect marker proteins with known PSLs without translocation events and proteins that can form similar distributions among fractions [ 12 ]. Notably, proteins with more than one PSL are not considered in marker proteins, similar to Mulvey et al.…”

“…’s [ 2 ] study. For each replicate in the experimental data, the data are normalized by obtaining the proportion of protein levels in each fraction [ 12 ]. We benchmark the performance of protein translocation methods using five simulated and six experimental spatio-temporal proteomics datasets ( Table 1 ).…”

“…The protein translocation labels in the synthetic dataset are used as object for supervised loss, whereas the expected probabilities of protein translocation labels in real dataset are used as object for unsupervised loss. To alleviate the effect of sample imbalance, the weighted cross entropy ( WCE ) loss function based on the number of samples in different classes [ 12 ] is used as the semi-supervised loss function, which is structured as Equation ( 9 ):…”

“…Image-based methods [ 11 ] such as immunohistochemistry and immunofluorescence can be employed to investigate protein subcellular localization (PSL) and translocation. However, it is time-consuming and requires specific antibodies and can be costly to study thousands of protein translocation events in different cell states [ 12 ]. Overexpression of tag-fused protein can avoid the use of protein-specific antibodies; however, certain cell lines are particularly difficult to transfect, such as the human monocytic leukemia cell line (THP-1) [ 2 ].…”