Machine Learning Strategy That Leverages Large Data sets to Boost Statistical Power in Small-Scale Experiments

STR mutations on chromosome 15q cause thyrotropin resistance by activating a primate-specific enhancer of MIR7-2/MIR1179

Thyrotropin (TSH) is the master regulator of thyroid gland growth and function. Resistance to TSH (RTSH) describes conditions with reduced sensitivity to TSH. Dominantly inherited RTSH has been linked to a locus on chromosome 15q, but its genetic basis has remained elusive. Here we show that non-coding mutations in a (TTTG)4 short tandem repeat (STR) underlie dominantly inherited RTSH in all 82 affected participants from 12 unrelated families. The STR is contained in a primate-specific Alu retrotransposon with thyroid-specific cis-regulatory chromatin features. Fiber-seq and RNA-seq studies revealed that the mutant STR activates a thyroid-specific enhancer cluster, leading to haplotype-specific upregulation of the bicistronic MIR7-2/MIR1179 locus 35 kb downstream and overexpression of its microRNA products in the participants' thyrocytes. An imbalance in signaling pathways targeted by these micro-RNAs provides a working model for this cause of RTSH. This finding broadens our current knowledge of genetic defects altering pituitary-thyroid feedback regulation.

RNxQuest: An Extension to the xQuest Pipeline Enabling Analysis of Protein-RNA Cross-Linking/Mass Spectrometry Data

Cross-linking and mass spectrometry (XL-MS) workflows are increasingly popular techniques for generating low-resolution structural information about interacting biomolecules. xQuest is an established software package for analysis of protein-protein XL-MS data, supporting stable isotope-labeled cross-linking reagents. Resultant paired peaks in mass spectra aid sensitivity and specificity of data analysis. The recently developed cross-linking of isotope-labeled RNA and mass spectrometry (CLIR-MS) approach extends the XL-MS concept to protein-RNA interactions, also employing isotope-labeled cross-link (XL) species to facilitate data analysis. Data from CLIR-MS experiments are broadly compatible with core xQuest functionality, but the required analysis approach for this novel data type presents several technical challenges not optimally served by the original xQuest package. Here we introduce RNxQuest, a Python package extension for xQuest, which automates the analysis approach required for CLIR-MS data, providing bespoke, state-of-the-art processing and visualization functionality for this novel data type. Using functions included with RNxQuest, we evaluate three false discovery rate control approaches for CLIR-MS data. We demonstrate the versatility of the RNxQuest-enabled data analysis pipeline by also reanalyzing published protein-RNA XL-MS data sets that lack isotope-labeled RNA. This study demonstrates that RNxQuest provides a sensitive and specific data analysis pipeline for detection of isotope-labeled XLs in protein-RNA XL-MS experiments.

Synchronized long-read genome, methylome, epigenome, and transcriptome for resolving a Mendelian condition

Resolving the molecular basis of a Mendelian condition (MC) remains challenging owing to the diverse mechanisms by which genetic variants cause disease. To address this, we developed a synchronized long-read genome, methylome, epigenome, and transcriptome sequencing approach, which enables accurate single-nucleotide, insertion-deletion, and structural variant calling and diploid de novo genome assembly, and permits the simultaneous elucidation of haplotype-resolved CpG methylation, chromatin accessibility, and full-length transcript information in a single long-read sequencing run. Application of this approach to an Undiagnosed Diseases Network (UDN) participant with a chromosome X;13 balanced translocation of uncertain significance revealed that this translocation disrupted the functioning of four separate genes (NBEA, PDK3, MAB21L1, and RB1) previously associated with single-gene MCs. Notably, the function of each gene was disrupted via a distinct mechanism that required integration of the four 'omes' to resolve. These included nonsense-mediated decay, fusion transcript formation, enhancer adoption, transcriptional readthrough silencing, and inappropriate X chromosome inactivation of autosomal genes. Overall, this highlights the utility of synchronized long-read multi-omic profiling for mechanistically resolving complex phenotypes.

Generalized precursor prediction boosts identification rates and accuracy in mass spectrometry based proteomics

Data independent acquisition mass spectrometry (DIA-MS) has recently emerged as an important method for the identification of blood-based biomarkers. However, the large search space required to identify novel biomarkers from the plasma proteome can introduce a high rate of false positives that compromise the accuracy of false discovery rates (FDR) using existing validation methods. We developed a generalized precursor scoring (GPS) method trained on 2.75 million precursors that can confidently control FDR while increasing the number of identified proteins in DIA-MS independent of the search space. We demonstrate how GPS can generalize to new data, increase protein identification rates, and increase the overall quantitative accuracy. Finally, we apply GPS to the identification of blood-based biomarkers and identify a panel of proteins that are highly accurate in discriminating between subphenotypes of septic acute kidney injury from undepleted plasma to showcase the utility of GPS in discovery DIA-MS proteomics.

Challenges and opportunities for single cell computational proteomics

Single-cell proteomics is growing rapidly and has made several technological advancements. As most research has been focused on improving instrumentation and sample preparation methods, very little attention has been given to algorithms responsible for identifying and quantifying proteins. Given the inherent difference between bulk data and single-cell data, it's necessary to realize that current algorithms being employed on single-cell data were designed for bulk data, and have underlying assumptions that may not hold true for single-cell data. In order to develop and optimize algorithms for single-cell data, we need to characterize the differences between single-cell data and bulk data, and assess how current algorithms perform on single-cell data. Here, we present a review of algorithms responsible for identifying and quantifying peptides and proteins. We will give a review of how each type of algorithm works, assumptions it relies on, how it performs on single-cell data, and possible optimizations and solutions that could be used to address the differences in single-cell data.

The Current State of Single-Cell Proteomics Data Analysis

Sound data analysis is essential to retrieve meaningful biological information from single-cell proteomics experiments. This analysis is carried out by computational methods that are assembled into workflows, and their implementations influence the conclusions that can be drawn from the data. In this work, we explore and compare the computational workflows that have been used over the last four years and identify a profound lack of consensus on how to analyze single-cell proteomics data. We highlight the need for benchmarking of computational workflows and standardization of computational tools and data, as well as carefully designed experiments. Finally, we cover the current standardization efforts that aim to fill the gap, list the remaining missing pieces, and conclude with lessons learned from the replication of published single-cell proteomics analyses. © 2023 Wiley Periodicals LLC.

Common Decoy Distributions Simplify False Discovery Rate Estimation in Shotgun Proteomics

In shotgun proteomics, false discovery rate (FDR) estimation is a necessary step to ensure the quality of accepted peptide-spectrum matches (PSMs) from a database search. Popular statistical validation tools for FDR control tend to rely on target-decoy searching to build empirical, dataset-specific models, which often leads to inaccurate FDR estimates. In this paper, we propose a new approach named common decoy distribution (CDD) to FDR estimation using the idea of a fixed empirical null score distribution derived from millions of peptide tandem mass spectra. To demonstrate the viability of CDD, its stability with respect to noise and the presence of unexpected peptide modifications was evaluated. PeptideProphet-based implementation of CDD was benchmarked against decoy-based PeptideProphet, and both methods exhibited similar accuracy of FDR estimates and retrieval of correct PSMs. The finding of this study calls for a re-evaluation of the necessity of dataset-specific target-decoy searches and illustrates the potential of Big Data approaches for statistical analysis in proteomics.

Driving Single Cell Proteomics Forward with Innovation

Current single-cell mass spectrometry (MS) methods can quantify thousands of peptides per single cell while detecting peptide-like features that may support the quantification of 10-fold more peptides. This 10-fold gain might be attained by innovations in data acquisition and interpretation even while using existing instrumentation. This perspective discusses possible directions for such innovations with the aim to stimulate community efforts for increasing the coverage and quantitative accuracy of single proteomics while simultaneously decreasing missing data. Parallel improvements in instrumentation, sample preparation, and peptide separation will afford additional gains. Together, these synergistic routes for innovation project a rapid growth in the capabilities of MS based single-cell protein analysis. These gains will directly empower applications of single-cell proteomics to biomedical research.

Deep learning boosts sensitivity of mass spectrometry-based immunopeptidomics

Characterizing the human leukocyte antigen (HLA) bound ligandome by mass spectrometry (MS) holds great promise for developing vaccines and drugs for immune-oncology. Still, the identification of non-tryptic peptides presents substantial computational challenges. To address these, we synthesized and analyzed >300,000 peptides by multi-modal LC-MS/MS within the ProteomeTools project representing HLA class I & II ligands and products of the proteases AspN and LysN. The resulting data enabled training of a single model using the deep learning framework Prosit, allowing the accurate prediction of fragment ion spectra for tryptic and non-tryptic peptides. Applying Prosit demonstrates that the identification of HLA peptides can be improved up to 7-fold, that 87% of the proposed proteasomally spliced HLA peptides may be incorrect and that dozens of additional immunogenic neo-epitopes can be identified from patient tumors in published data. Together, the provided peptides, spectra and computational tools substantially expand the analytical depth of immunopeptidomics workflows.

mokapot: Fast and Flexible Semisupervised Learning for Peptide Detection

Proteomics studies rely on the accurate assignment of peptides to the acquired tandem mass spectra-a task where machine learning algorithms have proven invaluable. We describe mokapot, which provides a flexible semisupervised learning algorithm that allows for highly customized analyses. We demonstrate some of the unique features of mokapot by improving the detection of RNA-cross-linked peptides from an analysis of RNA-binding proteins and increasing the consistency of peptide detection in a single-cell proteomics study.

Proteome Discoverer-A Community Enhanced Data Processing Suite for Protein Informatics

Proteomics researchers today face an interesting challenge: how to choose among the dozens of data processing and analysis pipelines available for converting tandem mass spectrometry files to protein identifications. Due to the dominance of Orbitrap technology in proteomics in recent history, many researchers have defaulted to the vendor software Proteome Discoverer. Over the fourteen years since the initial release of the software, it has evolved in parallel with the increasingly complex demands faced by proteomics researchers. Today, Proteome Discoverer exists in two distinct forms with both powerful commercial versions and fully functional free versions in use in many labs today. Throughout the 11 main versions released to date, a central theme of the software has always been the ability to easily view and verify the spectra from which identifications are made. This ability is, even today, a key differentiator from other data analysis solutions. In this review I will attempt to summarize the history and evolution of Proteome Discoverer from its first launch to the versions in use today.

Optimization of skeletal protein preparation for LC-MS/MS sequencing yields additional coral skeletal proteins in Stylophora pistillata

Stony corals generate their calcium carbonate exoskeleton in a highly controlled biomineralization process mediated by a variety of macromolecules including proteins. Fully identifying and classifying these proteins is crucial to understanding their role in exoskeleton formation, yet no optimal method to purify and characterize the full suite of extracted coral skeletal proteins has been established and hence their complete composition remains obscure. Here, we tested four skeletal protein purification protocols using acetone precipitation and ultrafiltration dialysis filters to present a comprehensive scleractinian coral skeletal proteome. We identified a total of 60 proteins in the coral skeleton, 44 of which were not present in previously published stony coral skeletal proteomes. Extracted protein purification protocols carried out in this study revealed that no one method captures all proteins and each protocol revealed a unique set of method-exclusive proteins. To better understand the general mechanism of skeletal protein transportation, we further examined the proteins’ gene ontology, transmembrane domains, and signal peptides. We found that transmembrane domain proteins and signal peptide secretion pathways, by themselves, could not explain the transportation of proteins to the skeleton. We therefore propose that some proteins are transported to the skeleton via non-traditional secretion pathways.

Single-cell protein analysis by mass spectrometry

Human physiology and pathology arise from the coordinated interactions of diverse single cells. However, analyzing single cells has been limited by the low sensitivity and throughput of analytical methods. DNA sequencing has recently made such analysis feasible for nucleic acids but single-cell protein analysis remains limited. Mass spectrometry is the most powerful method for protein analysis, but its application to single cells faces three major challenges: efficiently delivering proteins/peptides to mass spectrometry detectors, identifying their sequences, and scaling the analysis to many thousands of single cells. These challenges have motivated corresponding solutions, including SCoPE design multiplexing and clean, automated, and miniaturized sample preparation. Synergistically applied, these solutions enable quantifying thousands of proteins across many single cells and establish a solid foundation for further advances. Building upon this foundation, the SCoPE concept will enable analyzing subcellular organelles and posttranslational modifications, while increases in multiplexing capabilities will increase the throughput and decrease cost.