When it comes to proteomics data quality, the saying holds true: 'garbage in, garbage out.' But how do you spot garbage in LC-MS data? In this workshop, you will learn how to troubleshoot common proteomics sample preparation issues by taking a deep dive into LC-MS data profiles. How do I know if my samples are contaminated, overloaded, or poorly digested? Why do I have so few protein identifications in my LC-MS run? You will be challenged to examine LC-MS data quality using various software tools to answer these questions. With the help of instructors and your peers, you will learn to diagnose common mistakes in proteomics sample preparation and gain useful skills for your future experiments in the lab. You will also have the opporunity to discuss your own sample preparation questions and challenges arising from your own work. In these open discussion you can reived feedback and advice from experts in the field.

The assignment of tandem mass spectra to peptide sequences and the assembly to protein groups is arguably the key step in proteomics workflow, connecting experimental and quantitative data. Therefore, developers constantly aim to improve it and users constantly aim to get the best bang for their bucks.

In this workshop we will talk about the process of assigning spectra to peptides and proteins in more detail by investigating a (or multiple) commonly used database search engines together, from input to output. Please bring your own laptop – or team up with somebody!

In the workshop, we will cover the following topics:

1. Search parameters and their effects on identification rates
2. Databases and their effects on identification rates
3. Scoring systems and their implications on search spaces, sample complexity, and data quality
4. False discovery rate estimation on various levels and the implications when looking at large or multiple datasets
5. Post-processor rescoring (e.g. Percolator)
6. Data-driven rescoring (e.g. Prosit+Oktoberfest)
7. Practical implications of protein inference
8. Specifics in challenging datasets, e.g. metaproteomics, immunopeptidomics, post-translationally modified peptides
9. Differences between database searching and spectral library searching
10. Considerations when planning to use de novo sequencing

Where possible, we will be testing different settings and datasets to generate examples we will interactively discuss afterwards. The aim is to make you aware of the considerations when generating peptide and protein identification results. The topics we discuss here apply – to a very large extent – to any search engine you will apply since all these topics are generally applicable. Besides, what we will discuss, you will (hopefully) also learn how to easily and efficiently experiment with your data. This will not only allow you to get a feel for your data, but it will also teach you how to perform and interpret essential quality control checks on your acquired data.

Having said all this, there will also always be room for your own questions, issues, or challenges.

Data Independent Acquisition (DIA) allows for unbiased, robust quantification over numerous samples, providing deep proteome coverage and high quantitative sensitivity. directDIA, a spectrum centric approach for DIA data analysis, utilizes search space generated directly from DIA files. It has broad compatibility with analysis of tryptic peptides, semi-tryptic and completely unspecific ones, including advanced quantification of the PTM sites. These capabilities make directDIA a robust and convenient tool for DIA proteomics quantification.

In this workshop, we will highlight the benefits of using DIA for the accurate and reproducible quantification of proteins across biological samples. We will guide you through a library-free DIA analysis using Spectronaut directDIA workflow: from the data upload and choosing optimal settings through results inspection to results interpretation. An interactive environment will encourage discussions on relevant aspects of DIA analysis.

The workshop will cover:

1. Basics of DIA analysis and data processing strategies – 15 min

2. Hands-on data analysis using Spectronaut – 1.5 h

• Setting up data analysis using directDIA workflow
• Discussion of specific workflows – opportunities and potential pitfalls
• Discussion of settings selection – how to perform quantification for a given experimental strategy
• Analysis inspection and validation
• Results interpretation and statistics
• The importance of data quality for quantification

3. Q&A session, where workshop participants could ask questions related to the analysis of specific/own datasets – 15 min

For the hands-on workshop, the participants are required to bring a personal computer with these minimum requirements: Windows 10, x64, Intel® Core™ CPU, 2.7 GHz (4-cores) or similar, 200 GB free space, 8 GB of RAM, and .NET 8.0 preinstalled.

Registered participants will receive short licenses for Spectronaut, a demo dataset, and an analysis tutorial via email. Ideally, participants should have Spectronaut already installed on their computers before the workshop. The extended software licenses will enable participants to repeat exercises after the workshop or explore their own data after summer school.

By the end of the workshop, participants will have gained a thorough understanding of the intricacies involved in quantitative DIA analysis and will be equipped to effectively analyze DIA experiments using Spectronaut.

Regular maintenance and quality control (QC) are essential for maximising LC-MS uptime, ensuring data quality and supporting high-throughput workflows. In this interactive workshop, participants will explore how small issues can impact results and learn practical strategies for identifying and troubleshooting common problems. Using Skyline as a tool, participants will analyze real-case examples and recognize and diagnose typical LC-MS issues. By the end of the session, participants will be equipped with the knowledge to improve instrument performance and data quality through routine QC and maintenance practices.

Robust data quality assessment constitutes a critical prerequisite for obtaining meaningful and reproducible proteomic results. This workshop will address methodological aspects of data quality assessment in mass spectrometry-based proteomics, with a focus on moving beyond superficial summary metrics to examine low-level data interpretation and evaluation in proteomics workflows.

The program will cover the practical evaluation of raw data and chromatographic performance, critical analysis of extracted ion chromatograms (XICs) and fragment ion traces, quality assessment of MS/MS spectra, and strategies for external validation of false discovery rates (FDR) using entrapment and double-decoy approaches. Furthermore, the influence of data quality on peptide and protein identification rates will be discussed, alongside the limitations of coefficients of variation (CVs) as indicators of quantitative reproducibility.

The session will combine conceptual frameworks with practical examples and aims to provide an overview of current practices and challenges in data quality assessment beyond standard summary statistics. It is intended for researchers seeking to systematically evaluate their datasets and implement comprehensive quality assurance strategies within proteomics workflows

In this workshop, participants will engage in assignments that contrast structural proteomics methods with other structural biology techniques not based on mass spectrometry. The discussion will highlight the advantages and limitations of each approach, as well as the types of biological questions that can be addressed through MS-based structural analyses. The exercises will also delve into the biochemical principles underlying the LiP-MS approach introduced in the lecture. Additionally, time will be set aside to discuss how structural proteomics analyses can be tailored to address research questions of interest to the participants and to answer any related questions they may have.

Mass spectrometry (MS) has become an essential tool for proteome analysis of cellular compartments like mitochondria, of (single) cells or even tissues. It allows for the unbiased identification of proteins including their (post-translational) modifications in complex mixtures, as well as their relative and absolute quantitation. With recent methodological advances in sample preparation, MS instrumentation and data processing and evaluation, a nearly complete proteome of even highly complex biological systems can be achieved.

In addition to these global profiling approaches, protein MS also contributes to the area of cellular biochemistry and protein(-ligand) interaction analysis. Through the detection and relative quantification of protein-protein or protein-ligand interactions, it is possible to not only characterize discrete biomolecular complexes, but also entire networks and cellular pathways. This can be achieved by a range of approaches including affinity purification, co-fractionation by near-native gel electrophoresis or size exclusion chromatography (complexome profiling), proximity labelling of interacting proteins or (photo-)chemical cross-linking.

In this workshop I will cover the above protein interaction techniques. I will give examples together with practical considerations for affinity purification mass spectrometry (AP-MS), which is still the most widely used approach in MS-based protein interaction analysis. I will explain so-called complexome profiling and its applications in monitoring protein complexes derived from cells, organelles and/or tissues on a global level. Furthermore, some aspects of proximity labelling approaches will be discussed. Finally, special emphasis will be put on protein-protein and protein-RNA/DNA crosslinking of isolated complexes and entire cells. 

Cells respond to environmental cues, such as signals or stresses, by dynamically remodelling their proteome. A great part of the proteome-level regulation happens post-translationally. Much of the post-translational regulation involves chemical modifications in the protein, although there are other mechanisms that can control protein function at this level, such as subcellular relocation, secretion or targeted degradation.

Mass spectrometry-based proteomics offers an exceptional platform to explore from a systems-level perspective the dynamics involved in PTMs-driven regulation, allowing to identify these modifications at site resolution. However, the PTM landscape is extremely wide, including enzymatic and non-enzymatic modifications: phosphorylation, acetylation, methylation, ubiquitylation, SUMOylation, glycosylation, oxidations, among many others.

In this workshop we will cover the main challenges on PTM analysis, both from the experimental and the data analysis perspective by doing hands-on examples on real data. By doing so, we will learn how to explore databases such as Phosphositeplus (1) or Signor (2), employ tools for visualization such as OmicsVisualizer (3), infer kinase activity using RoKai (4) or perform exploratory analysis of the global PTM universe by doing open-search strategies (5).

We recommend all participants to bring their own laptops. It is highly advisable for all participants to have the following tools installed: Perseus (https://maxquant.net/perseus/), Perseus Plugin for Peptide Collapse (https://github.com/AlexHgO/Perseus_Plugin_Peptide_Collapse), Cytoscape (https://cytoscape.org/ including the following two apps - https://apps.cytoscape.org/apps/signorapp and https://apps.cytoscape.org/apps/omicsvisualizer) and FragPipe (https://fragpipe.nesvilab.org/).

Recommended literature:

1. Hornbeck PV et al.  PhosphoSitePlus, 2014: mutations, PTMs and recalibrations. Nucleic Acids Res. 2015 43:D512-20
2. Lo Surdo P, et al. SIGNOR 3.0, the SIGnaling network open resource 3.0: 2022 update. Nucleic Acids Res. 2022 Oct 16:gkac883.
3. Legeay M, et al. Visualize omics data on networks with Omics Visualizer, a Cytoscape App. F1000Res. 2020 Feb 28;9:157.
4. Yilmaz S, et al. (2021) Robust inference of kinase activity using functional networks Nature Communications 12 (1117)
5. Chick JM, et al. A mass-tolerant database search identifies a large proportion of unassigned spectra in shotgun proteomics as modified peptides. Nat Biotechnol. 2015 Jul;33(7):743-9.
6. Martinez-Val A, et al. Urea Artifacts Interfere with Immuno-Purification of Lysine Acetylation. J Proteome Res. 2017 Feb 3;16(2):1061-1068. doi: 10.1021/acs.jproteome.6b00463. Epub 2017 Jan 17. PMID: 28067524.
7. Martínez-Val A, et al. Hybrid-DIA: intelligent data acquisition integrates targeted and discovery proteomics to analyze phospho-signaling in single spheroids. Nat Commun. 2023 Jun 16;14(1):3599. doi: 10.1038/s41467-023-39347-y. PMID: 37328457; PMCID: PMC10276052.
8. Franciosa G, et al. Recent advances in kinase signaling network profiling by mass spectrometry. Curr Opin Chem Biol. 2023 Apr;73:102260. doi: 10.1016/j.cbpa.2022.102260. Epub 2023 Jan 18. PMID: 36657259
9. Ochoa D, et al. The functional landscape of the human phosphoproteome. Nat Biotechnol. 2020 Mar;38(3):365-373. doi: 10.1038/s41587-019-0344-3.
10. Leutert M, et al. Decoding Post-Translational Modification Crosstalk With Proteomics. Mol Cell Proteomics. 2021;20:100129.

Spatial proteomics (SP) offers unprecedented insights into the localization, quantity, and interactions of proteins within cells and tissues. Methods to study the spatially resolved proteome are diverse and vary in scale, resolution, sensitivity, and cellular throughput. Broadly, they can be subdivided into targeted, antibody-based, and untargeted mass spectrometry-based approaches. Owing to their highly complementary nature, multiscale spatial proteomics methods have been developed more recently, which aim at leveraging the complementary strengths of targeted and untargeted SP. One example is deep visual proteomics (DVP), which combines whole-slide imaging, machine-learning-based image analysis, automated laser microdissection, and ultrasensitive MS-based proteomics. This emerging discovery proteomics approach is under continuous development to advance throughput, scalability, robustness, and versatility. 

In this workshop, you will learn practical tips and tricks on how to perform spatial proteomics experiments based on multiscale approaches, such as deep visual proteomics. You do not need any experience with spatial proteomics, but ideally are interested in applying this method for your own research. Maybe you have already performed experiments and want to bring up your own ideas/challenges/discussion points to the workshop?

We cover and discuss the following topics:

1. Overview of hardware and software requirements, discussion on advantages and disadvantages
2. The versatility and challenges of laser microdissection workflows – how do I get started and what are often faced problems?
3. Low-input tissue proteomics: How to perform sample preparation? What are the most critical steps?
4. Considerations for experimental design (Which spatial resolution? ROI or single-cell profiling? Which mass spectrometer do I need?)
5. Learn a new, fast-track DVP workflow that does not require prior image analysis experience
6. Optional: Troubleshooting Q&A to discuss problems and practical solutions

What you should bring:

Standard laptop, please install the freely available image analysis software QuPath: https://qupath.readthedocs.io/en/0.5/. We will discuss a simple workflow to guide MS-based proteomics by microscopy.