Dataset Information

Direct data-independent acquisition proteomics with in silico spectral libraries by deep learning

ABSTRACT: Data-independent acquisition (DIA) proteomics allows systematic and unbiased measurement of protein samples and enables fast quantitative analysis of large cohorts of samples. However, sample-specific spectral libraries are usually required prior to perform DIA experiments. The libraries are built by data-dependent acquisition (DDA) proteomic analysis on the same samples normally with pre-fractionation, which is time-consuming and limits the identification/quantification by DIA to the peptides identified by DDA. Herein, we propose DeepDIA, a deep learning based approach to generate in silico spectral libraries for DIA analysis. We benchmarked DeepDIA with HeLa and mixed proteome data sets, showing that the quality of in silico spectral libraries is comparable to that of experimental spectral libraries. We further demonstrated that DeepDIA can be performed on human cell lines and human serum without pre-knowledge of peptides lists by DDA experiments. Compared to the state-of-the-art protocol using DDA-based spectral library with high abundance protein depletion and pre-fractionation, the number of identified and quantified proteins from human serum was increased by >100% using DeepDIA. Accuracy of the identification results was validated using a standard mixture containing >800 stable isotope labelled reference peptides from >500 proteins in human plasma. We expect this work contributing to the studies of quantitative proteomics and especially blood proteomics, whereas expanding the toolbox for DIA proteomics.

ORGANISM(S): Homo Sapiens

SUBMITTER: Liang Qiao

PROVIDER: PXD014108 | iProX | Mon Dec 02 00:00:00 GMT 2019

REPOSITORIES: iProX

ACCESS DATA

Publications

In silico spectral libraries by deep learning facilitate data-independent acquisition proteomics.

Yang Yi Y Liu Xiaohui X Shen Chengpin C Lin Yu Y Yang Pengyuan P Qiao Liang L

Nature communications 20200109 1

Data-independent acquisition (DIA) is an emerging technology for quantitative proteomic analysis of large cohorts of samples. However, sample-specific spectral libraries built by data-dependent acquisition (DDA) experiments are required prior to DIA analysis, which is time-consuming and limits the identification/quantification by DIA to the peptides identified by DDA. Herein, we propose DeepDIA, a deep learning-based approach to generate in silico spectral libraries for DIA analysis. We demonstr ...[more]

PMID: 31919359

Similar Datasets

Project description:Data independent acquisition (DIA or DIA/SWATH ) mass spectrometry has emerged as a primary measurement strategy in the field of quantitative proteomics. diaPASEF is a recent adaptation that leverages trapped ion mobility spectrometry (TIMS) to improve selectivity and increase sensitivity. The complex fragmentation spectra generated by co-isolation of peptides in DIA mode are most typically analyzed with reference to prior knowledge in the form of spectral libraries. The best established method for generating libraries uses data dependent acquisition (DDA) mode, or DIA mode if appropriately deconvoluted, often including offline fractionation to increase depth of coverage,to create spectral libraries. More recently strategies for spectral library generation based on gas phase fractionation (GPF), where a representative sample is injected serially using narrow window DIA methods designed to cover different slices of the precursor space, have been introduced and performed comparably to deep offline fractionation-based libraries for DIA data analysis. Here, we investigated whether an analogous GPF-based library building approach that accounts for the ion mobility (IM) dimension is useful for the analysis of diaPASEF data and can remove the need for offline fractionation. To enable a rapid library development approach for diaPASEF we designed a GPF acquisition scheme covering the majority of multiply charged precursors in the m/z vs 1/K0 space requiring 7 injections of a representative sample and compared this with libraries generated by direct deconvolution-based analysis of diaPASEF data or by deep offline fractionation and ddaPASEF. . We found that the GPF based library outperformed library generation by direct deconvolution of the diaPASEF data, and performed comparably to deep offline fractionation libraries, when analysing diaPASEF data acquired from 200ng of commercial HeLa digest. With the ion mobility integrated GPF scheme we establish a pragmatic approach to rapid and comprehensive library generation for the analysis of diaPASEF data.

Dataset Information

Direct data-independent acquisition proteomics with in silico spectral libraries by deep learning

Publications

In silico spectral libraries by deep learning facilitate data-independent acquisition proteomics.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure