Search

Machine learning and in particular deep learning (DL) are increasingly important in mass spectrometry (MS)-based proteomics. Recent DL models can predict the retention time, ion mobility and fragment intensities of a peptide just from the amino acid sequence with good accuracy. However, DL is a very rapidly developing field with new neural network architectures frequently appearing, which are challenging to incorporate for proteomics researchers. Here we introduce AlphaPeptDeep, a modular Python framework built on the PyTorch DL library that learns and predicts the properties of peptides (https://github.com/MannLabs/alphapeptdeep). It features a model shop that enables non-specialists to create models in just a few lines of code. AlphaPeptDeep represents post-translational modifications in a generic manner, even if only the chemical composition is known. Extensive use of transfer learning obviates the need for large data sets to refine models for particular experimental conditions. The AlphaPeptDeep models for predicting retention time, collisional cross sections and fragment intensities are at least on par with existing tools. Additional sequence-based properties can also be predicted by AlphaPeptDeep, as demonstrated with a HLA peptide prediction model to improve HLA peptide identification for data-independent acquisition (https://github.com/MannLabs/PeptDeep-HLA).

Single-cell technologies are revolutionizing biology but are today mainly limited to imaging and deep sequencing1–3. However, proteins are the main drivers of cellular function and in-depth characterization of individual cells by mass spectrometry (MS)-based proteomics would thus be highly valuable and complementary4,5. Chemical labeling-based single-cell approaches introduce hundreds of cells into the MS, but direct analysis of single cells has not yet reached the necessary sensitivity, robustness and quantitative accuracy to answer biological questions6,7. Here, we develop a robust workflow combining miniaturized sample preparation, very low flow-rate chromatography and a novel trapped ion mobility mass spectrometer, resulting in a more than ten-fold improved sensitivity. We accurately and robustly quantify proteomes and their changes in single, FACS-isolated cells. Arresting cells at defined stages of the cell cycle by drug treatment retrieves expected key regulators such as CDK2NA, the E2 ubiquitin ligase UBE2S, DNA topoisomerases TOP2A/B and the chromatin regulator HMGA1. Furthermore, it highlights potential novel ones and allows cell phase prediction. Comparing the variability in more than 430 single-cell proteomes to transcriptome data revealed a stable core proteome despite perturbation, while the transcriptome appears volatile. This emphasizes substantial regulation of translation and sets the stage for its elucidation at the single cell level. Our technology can readily be applied to ultra-high sensitivity analyses of tissue material8, posttranslational modifications and small molecule studies to gain unprecedented insights into cellular heterogeneity in health and disease.

Mass spectrometry based proteomics is the method of choice for quantifying genome-wide differential changes of protein expression in a wide range of biological and biomedical applications. Protein expression changes need to be reliably derived from many measured peptide intensities and their corresponding peptide fold changes. These peptide fold changes vary considerably for a given protein. Numerous instrumental setups aim to reduce this variability, whereas current computational methods only implicitly account for this problem. We introduce a new method, MS-EmpiRe, which explicitly accounts for the noise underlying peptide fold changes. We derive data set-specific, intensity-dependent empirical error fold change distributions, which are used for individual weighing of peptide fold changes to detect differentially expressed proteins (DEPs). In a recently published proteome-wide benchmarking data set, MS-EmpiRe doubles the number of correctly identified DEPs at an estimated FDR cutoff compared with state-of-the-art tools. We additionally confirm the superior performance of MS-EmpiRe on simulated data. MS-EmpiRe requires only peptide intensities mapped to proteins and, thus, can be applied to any common quantitative proteomics setup. We apply our method to diverse MS data sets and observe consistent increases in sensitivity with more than 1000 additional significant proteins in deep data sets, including a clinical study over multiple patients. At the same time, we observe that even the proteins classified as most insignificant by other methods but significant by MS-EmpiRe show very clear regulation on the peptide intensity level. MS-EmpiRe provides rapid processing (< 2 min for 6 LC-MS/MS runs (3 h gradients)) and is publicly available under github.com/zimmerlab/MS-EmpiRe with a manual including examples.

Spectral libraries play a central role in the analysis of data independent acquisition (DIA) proteomics experiments. DIA experiments require spectral libraries, as most current methods cannot apply traditional peptide identification via database searching on DIA data. A central assumption in current spectral library tools is that a single characteristic intensity pattern (CIP) suffices to describe the fragmentation of an unmodified peptide in a particular parent ion charge state (peptide charge pair). However, we find that this is often not the case. We analyze a heterogeneous dataset of 440.000 MaxQuant - preprocessed peptide spectra from a QToF mass spectrometer, stemming from over 100 different LC-MS/MS runs. The dataset corresponds to 10.580 peptide charge pairs, which have each been measured and identified at least 20 times. We demonstrate that the same charged and unmodified peptide can fragment in multiple reproducible ways, even within the same LC-MS/MS run. We integrate multiple reference CIPs (MCIPs) in our model library and observe a >99% coverage of replicate fragmentation spectra for 95% of peptide charge pairs (using up to four CIPs). Using a single CIP (as in current spectral library approaches), we find >99% coverage for only 50% of the peptide charge pairs. Our approach achieves substantially greater sensitivity in comparison to the popular SpectraST library generation tool. Using randomized decoy spectra, we demonstrate that identification accuracy of the MCIP approach is improved by up to 12% compared to a single CIP approach. We test the MCIP approach on a SWATH data set and observe a ~30% increase in peptide recognition. We conclude that including MCIPs in spectral library approaches would yield increased sensitivity without compromising the false discovery rate.