Trans-omic profiling uncovers molecular controls of early human cerebral organoid formation.

Defining the molecular networks orchestrating human brain formation is crucial for understanding neurodevelopment and neurological disorders. Challenges in acquiring early brain tissue have incentivized the use of three-dimensional human pluripotent stem cell (hPSC)-derived neural organoids to recapitulate neurodevelopment. To elucidate the molecular programs that drive this highly dynamic process, here, we generate a comprehensive trans-omic map of the phosphoproteome, proteome, and transcriptome of the exit of pluripotency and neural differentiation toward human cerebral organoids (hCOs). These data reveal key phospho-signaling events and their convergence on transcriptional factors to regulate hCO formation. Comparative analysis with developing human and mouse embryos demonstrates the fidelity of our hCOs in modeling embryonic brain development. Finally, we demonstrate that biochemical modulation of AKT signaling can control hCO differentiation. Together, our data provide a comprehensive resource to study molecular controls in human embryonic brain development and provide a guide for the future development of hCO differentiation protocols.

Trafficking regulator of GLUT4-1 (TRARG1) is a GSK3 substrate.

Trafficking regulator of GLUT4-1, TRARG1, positively regulates insulin-stimulated GLUT4 trafficking and insulin sensitivity. However, the mechanism(s) by which this occurs remain(s) unclear. Using biochemical and mass spectrometry analyses we found that TRARG1 is dephosphorylated in response to insulin in a PI3K/Akt-dependent manner and is a novel substrate for GSK3. Priming phosphorylation of murine TRARG1 at serine 84 allows for GSK3-directed phosphorylation at serines 72, 76 and 80. A similar pattern of phosphorylation was observed in human TRARG1, suggesting that our findings are translatable to human TRARG1. Pharmacological inhibition of GSK3 increased cell surface GLUT4 in cells stimulated with a submaximal insulin dose, and this was impaired following Trarg1 knockdown, suggesting that TRARG1 acts as a GSK3-mediated regulator in GLUT4 trafficking. These data place TRARG1 within the insulin signaling network and provide insights into how GSK3 regulates GLUT4 trafficking in adipocytes.

Functional analysis of the stable phosphoproteome reveals cancer vulnerabilities.

MOTIVATION: The advance of mass spectrometry-based technologies enabled the profiling of the phosphoproteomes of a multitude of cell and tissue types. However, current research primarily focused on investigating the phosphorylation dynamics in specific cell types and experimental conditions, whereas the phosphorylation events that are common across cell/tissue types and stable regardless of experimental conditions are, so far, mostly ignored. RESULTS: Here, we developed a statistical framework to identify the stable phosphoproteome across 53 human phosphoproteomics datasets, covering 40 cell/tissue types and 194 conditions/treatments. We demonstrate that the stably phosphorylated sites (SPSs) identified from our statistical framework are evolutionarily conserved, functionally important and enriched in a range of core signaling and gene pathways. Particularly, we show that SPSs are highly enriched in the RNA splicing pathway, an essential cellular process in mammalian cells, and frequently disrupted by cancer mutations, suggesting a link between the dysregulation of RNA splicing and cancer development through mutations on SPSs. AVAILABILITY AND IMPLEMENTATION: The source code for data analysis in this study is available from Github repository https://github.com/PYangLab/SPSs under the open-source license of GPL-3. The data used in this study are publicly available (see Section 2.8). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Decoding the function of bivalent chromatin in development and cancer.

Bivalent chromatin is characterized by the simultaneous presence of H3K4me3 and H3K27me3, histone modifications generally associated with transcriptionally active and repressed chromatin, respectively. Prevalent in embryonic stem cells (ESCs), bivalency is postulated to poise/prime lineage-controlling developmental genes for rapid activation during embryogenesis while maintaining a transcriptionally repressed state in the absence of activation cues; however, this hypothesis remains to be directly tested. Most gene promoters DNA hypermethylated in adult human cancers are bivalently marked in ESCs, and it was speculated that bivalency predisposes them for aberrant de novo DNA methylation and irreversible silencing in cancer, but evidence supporting this model is largely lacking. Here, we show that bivalent chromatin does not poise genes for rapid activation but protects promoters from de novo DNA methylation. Genome-wide studies in differentiating ESCs reveal that activation of bivalent genes is no more rapid than that of other transcriptionally silent genes, challenging the premise that H3K4me3 is instructive for transcription. H3K4me3 at bivalent promoters-a product of the underlying DNA sequence-persists in nearly all cell types irrespective of gene expression and confers protection from de novo DNA methylation. Bivalent genes in ESCs that are frequent targets of aberrant hypermethylation in cancer are particularly strongly associated with loss of H3K4me3/bivalency in cancer. Altogether, our findings suggest that bivalency protects reversibly repressed genes from irreversible silencing and that loss of H3K4me3 may make them more susceptible to aberrant DNA methylation in diseases such as cancer. Bivalency may thus represent a distinct regulatory mechanism for maintaining epigenetic plasticity.

Akt phosphorylates insulin receptor substrate to limit PI3K-mediated PIP3 synthesis.

The phosphoinositide 3-kinase (PI3K)-Akt network is tightly controlled by feedback mechanisms that regulate signal flow and ensure signal fidelity. A rapid overshoot in insulin-stimulated recruitment of Akt to the plasma membrane has previously been reported, which is indicative of negative feedback operating on acute timescales. Here, we show that Akt itself engages this negative feedback by phosphorylating insulin receptor substrate (IRS) 1 and 2 on a number of residues. Phosphorylation results in the depletion of plasma membrane-localised IRS1/2, reducing the pool available for interaction with the insulin receptor. Together these events limit plasma membrane-associated PI3K and phosphatidylinositol (3,4,5)-trisphosphate (PIP3) synthesis. We identified two Akt-dependent phosphorylation sites in IRS2 at S306 (S303 in mouse) and S577 (S573 in mouse) that are key drivers of this negative feedback. These findings establish a novel mechanism by which the kinase Akt acutely controls PIP3 abundance, through post-translational modification of the IRS scaffold.

For the body to work properly, cells must constantly 'talk' to each other using signalling molecules. Receiving a chemical signal triggers a series of molecular events in a cell, a so-called 'signal transduction pathway' that connects a signal with a precise outcome. Disturbing cell signalling can trigger disease, and strict control mechanisms are therefore in place to ensure that communication does not break down or become erratic. For instance, just as a thermostat turns off the heater once the right temperature is reached, negative feedback mechanisms in cells switch off signal transduction pathways when the desired outcome has been achieved. The hormone insulin is a signal for growth that increases in the body following a meal to promote the storage of excess blood glucose (sugar) in muscle and fat cells. The hormone binds to insulin receptors at the cell surface and switches on a signal transduction pathway that makes the cell take up glucose from the bloodstream. If the signal is not engaged diseases such as diabetes develop. Conversely, if the signal cannot be adequately switched of cancer can develop. Determining exactly how insulin works would help to understand these diseases better and to develop new treatments. Kearney et al. therefore set out to examine the biochemical 'fail-safes' that control insulin signalling. Experiments using computer simulations of the insulin signalling pathway revealed a potential new mechanism for negative feedback, which centred on a molecule known as Akt. The models predicted that if the negative feedback were removed, then Akt would become hyperactive and accumulate at the cell's surface after stimulation with insulin. Further manipulation of the 'virtual' insulin signalling pathway and studies of live cells in culture confirmed that this was indeed the case. The cell biology experiments also showed how Akt, once at the cell surface, was able to engage the negative feedback and shut down further insulin signalling. Akt did this by inactivating a protein required to pass the signal from the insulin receptor to the rest of the cell. Overall, this work helps to understand cell communication by revealing a previously unknown, and critical component of the insulin signalling pathway.

Global phosphoproteomics reveals DYRK1A regulates CDK1 activity in glioblastoma cells.

Both tumour suppressive and oncogenic functions have been reported for dual-specificity tyrosine phosphorylation-regulated kinase 1A (DYRK1A). Herein, we performed a detailed investigation to delineate the role of DYRK1A in glioblastoma. Our phosphoproteomic and mechanistic studies show that DYRK1A induces degradation of cyclin B by phosphorylating CDC23, which is necessary for the function of the anaphase-promoting complex, a ubiquitin ligase that degrades mitotic proteins. DYRK1A inhibition leads to the accumulation of cyclin B and activation of CDK1. Importantly, we established that the phenotypic response of glioblastoma cells to DYRK1A inhibition depends on both retinoblastoma (RB) expression and the degree of residual DYRK1A activity. Moderate DYRK1A inhibition leads to moderate cyclin B accumulation, CDK1 activation and increased proliferation in RB-deficient cells. In RB-proficient cells, cyclin B/CDK1 activation in response to DYRK1A inhibition is neutralized by the RB pathway, resulting in an unchanged proliferation rate. In contrast, complete DYRK1A inhibition with high doses of inhibitors results in massive cyclin B accumulation, saturation of CDK1 activity and cell cycle arrest, regardless of RB status. These findings provide new insights into the complexity of context-dependent DYRK1A signalling in cancer cells.

Clonal evolution in liver cancer at single-cell and single-variant resolution.

Genetic heterogeneity of tumor is closely related to its clonal evolution, phenotypic diversity and treatment resistance, and such heterogeneity has only been characterized at single-cell sub-chromosomal scale in liver cancer. Here we reconstructed the single-variant resolution clonal evolution in human liver cancer based on single-cell mutational profiles. The results indicated that key genetic events occurred early during tumorigenesis, and an early metastasis followed by independent evolution was observed in primary liver tumor and intrahepatic metastatic portal vein tumor thrombus. By parallel single-cell RNA-Seq, the transcriptomic phenotype of HCC was found to be related with genetic heterogeneity. For the first time we reconstructed the single-cell and single-variant clonal evolution in human liver cancer, and dissection of both genetic and phenotypic heterogeneity will facilitate better understanding of their relationship.

Integrated Glycoproteomics Identifies a Role of N-Glycosylation and Galectin-1 on Myogenesis and Muscle Development.

Many cell surface and secreted proteins are modified by the covalent addition of glycans that play an important role in the development of multicellular organisms. These glycan modifications enable communication between cells and the extracellular matrix via interactions with specific glycan-binding lectins and the regulation of receptor-mediated signaling. Aberrant protein glycosylation has been associated with the development of several muscular diseases, suggesting essential glycan- and lectin-mediated functions in myogenesis and muscle development, but our molecular understanding of the precise glycans, catalytic enzymes, and lectins involved remains only partially understood. Here, we quantified dynamic remodeling of the membrane-associated proteome during a time-course of myogenesis in cell culture. We observed wide-spread changes in the abundance of several important lectins and enzymes facilitating glycan biosynthesis. Glycomics-based quantification of released N-linked glycans confirmed remodeling of the glycome consistent with the regulation of glycosyltransferases and glycosidases responsible for their formation including a previously unknown digalactose-to-sialic acid switch supporting a functional role of these glycoepitopes in myogenesis. Furthermore, dynamic quantitative glycoproteomic analysis with multiplexed stable isotope labeling and analysis of enriched glycopeptides with multiple fragmentation approaches identified glycoproteins modified by these regulated glycans including several integrins and growth factor receptors. Myogenesis was also associated with the regulation of several lectins, most notably the upregulation of galectin-1 (LGALS1). CRISPR/Cas9-mediated deletion of Lgals1 inhibited differentiation and myotube formation, suggesting an early functional role of galectin-1 in the myogenic program. Importantly, similar changes in N-glycosylation and the upregulation of galectin-1 during postnatal skeletal muscle development were observed in mice. Treatment of new-born mice with recombinant adeno-associated viruses to overexpress galectin-1 in the musculature resulted in enhanced muscle mass. Our data form a valuable resource to further understand the glycobiology of myogenesis and will aid the development of intervention strategies to promote healthy muscle development or regeneration.

PhosR enables processing and functional analysis of phosphoproteomic data.

Mass spectrometry (MS)-based phosphoproteomics has revolutionized our ability to profile phosphorylation-based signaling in cells and tissues on a global scale. To infer the action of kinases and signaling pathways in phosphoproteomic experiments, we present PhosR, a set of tools and methodologies implemented in a suite of R packages facilitating comprehensive analysis of phosphoproteomic data. By applying PhosR to both published and new phosphoproteomic datasets, we demonstrate capabilities in data imputation and normalization by using a set of "stably phosphorylated sites" and in functional analysis for inferring active kinases and signaling pathways. In particular, we introduce a "signalome" construction method for identifying a collection of signaling modules to summarize and visualize the interaction of kinases and their collective actions on signal transduction. Together, our data and findings demonstrate the utility of PhosR in processing and generating biological knowledge from MS-based phosphoproteomic data.

Data Independent Acquisition Proteomic Analysis Can Discriminate between Actinic Keratosis, Bowen's Disease, and Cutaneous Squamous Cell Carcinoma.

Actinic keratosis, Bowen's disease and cutaneous squamous cell carcinoma (cSCC) are heterogeneous keratinocytic skin lesions. Biomarkers that can accurately stratify these lesion types are needed to support a new paradigm of personalized and precise management of skin neoplasia. In this paper, we used a data independent acquisition proteomics workflow, sequential window acquisition of all theoretical mass spectra, to analyze formalin-fixed paraffin-embedded samples of normal skin and keratinocytic skin lesions, including well-differentiated, moderately differentiated and poorly differentiated cSCC lesions. We quantified 3,574 proteins across the 93 samples studied. Differential abundance analysis identified 19, 5, and 6 protein markers exclusive to actinic keratosis, Bowen's disease and cSCC lesions, respectively. Among cSCC lesions of various levels of tumor differentiation, 118, 230, and 17 proteins showed a potential as biomarkers of well-differentiated, moderately differentiated and poorly differentiated cSCC lesions, respectively. Bioinformatics analysis revealed that actinic keratosis and cSCC lesions were associated with decreased apoptosis, and Bowen's disease lesions with over-representation of the DNA damage repair pathway. Differential expression of alternatively spliced FGFR2, Rho guanosine triphosphatase signaling, and RNA metabolism proteins were associated with the level of cSCC tumor differentiation. Proteome profiles also separated keratinocytic skin lesion subtypes on principal components analysis. Overall, protein markers have excellent potential to discriminate keratinocytic skin lesion subtypes and facilitate new diagnostic and therapeutic strategies.

scReClassify: post hoc cell type classification of single-cell rNA-seq data.

BACKGROUND: Single-cell RNA-sequencing (scRNA-seq) is a fast emerging technology allowing global transcriptome profiling on the single cell level. Cell type identification from scRNA-seq data is a critical task in a variety of research such as developmental biology, cell reprogramming, and cancers. Typically, cell type identification relies on human inspection using a combination of prior biological knowledge (e.g. marker genes and morphology) and computational techniques (e.g. PCA and clustering). Due to the incompleteness of our current knowledge and the subjectivity involved in this process, a small amount of cells may be subject to mislabelling. RESULTS: Here, we propose a semi-supervised learning framework, named scReClassify, for 'post hoc' cell type identification from scRNA-seq datasets. Starting from an initial cell type annotation with potentially mislabelled cells, scReClassify first performs dimension reduction using PCA and next applies a semi-supervised learning method to learn and subsequently reclassify cells that are likely mislabelled initially to the most probable cell types. By using both simulated and real-world experimental datasets that profiled various tissues and biological systems, we demonstrate that scReClassify is able to accurately identify and reclassify misclassified cells to their correct cell types. CONCLUSIONS: scReClassify can be used for scRNA-seq data as a post hoc cell type classification tool to fine-tune cell type annotations generated by any cell type classification procedure. It is implemented as an R package and is freely available from https://github.com/SydneyBioX/scReClassify.

Global redox proteome and phosphoproteome analysis reveals redox switch in Akt.

Protein oxidation sits at the intersection of multiple signalling pathways, yet the magnitude and extent of crosstalk between oxidation and other post-translational modifications remains unclear. Here, we delineate global changes in adipocyte signalling networks following acute oxidative stress and reveal considerable crosstalk between cysteine oxidation and phosphorylation-based signalling. Oxidation of key regulatory kinases, including Akt, mTOR and AMPK influences the fidelity rather than their absolute activation state, highlighting an unappreciated interplay between these modifications. Mechanistic analysis of the redox regulation of Akt identified two cysteine residues in the pleckstrin homology domain (C60 and C77) to be reversibly oxidized. Oxidation at these sites affected Akt recruitment to the plasma membrane by stabilizing the PIP3 binding pocket. Our data provide insights into the interplay between oxidative stress-derived redox signalling and protein phosphorylation networks and serve as a resource for understanding the contribution of cellular oxidation to a range of diseases.

QCMAP: An Interactive Web-Tool for Performance Diagnosis and Prediction of LC-MS Systems.

The increasing role played by liquid chromatography-mass spectrometry (LC-MS)-based proteomics in biological discovery has led to a growing need for quality control (QC) on the LC-MS systems. While numerous quality control tools have been developed to track the performance of LC-MS systems based on a pre-defined set of performance factors (e.g., mass error, retention time), the precise influence and contribution of the performance factors and their generalization property to different biological samples are not as well characterized. Here, a web-based application (QCMAP) is developed for interactive diagnosis and prediction of the performance of LC-MS systems across different biological sample types. Leveraging on a standardized HeLa cell sample run as QC within a multi-user facility, predictive models are trained on a panel of commonly used performance factors to pinpoint the precise conditions to a (un)satisfactory performance in three LC-MS systems. It is demonstrated that the learned model can be applied to predict LC-MS system performance for brain samples generated from an independent study. By compiling these predictive models into our web-application, QCMAP allows users to benchmark the performance of their LC-MS systems using their own samples and identify key factors for instrument optimization. QCMAP is freely available from: http://shiny.maths.usyd.edu.au/QCMAP/.

An improved Akt reporter reveals intra- and inter-cellular heterogeneity and oscillations in signal transduction.

Akt is a key node in a range of signal transduction cascades and play a critical role in diseases such as cancer and diabetes. Fluorescently-tagged Akt reporters have been used to discern Akt localisation, yet it has not been clear how well these tools recapitulate the behaviour of endogenous Akt proteins. Here, we observed that fusion of eGFP to Akt2 impaired both its insulin-stimulated plasma membrane recruitment and its phosphorylation. Endogenous-like responses were restored by replacing eGFP with TagRFP-T. The improved response magnitude and sensitivity afforded by TagRFP-T-Akt2 over eGFP-Akt2 enabled monitoring of signalling outcomes in single cells at physiological doses of insulin with subcellular resolution and revealed two previously unreported features of Akt biology. In 3T3-L1 adipocytes, stimulation with insulin resulted in recruitment of Akt2 to the plasma membrane in a polarised fashion. Additionally, we observed oscillations in plasma membrane localised Akt2 in the presence of insulin with a consistent periodicity of 2 min. Our studies highlight the importance of fluorophore choice when generating reporter constructs and shed light on new Akt signalling responses that may encode complex signalling information.This article has an associated First Person interview with the first author of the paper.

mTORC1 Is a Major Regulatory Node in the FGF21 Signaling Network in Adipocytes.

FGF21 improves the metabolic profile of obese animals through its actions on adipocytes. To elucidate the signaling network responsible for mediating these effects, we quantified dynamic changes in the adipocyte phosphoproteome following acute exposure to FGF21. FGF21 regulated a network of 821 phosphosites on 542 proteins. A major FGF21-regulated signaling node was mTORC1/S6K. In contrast to insulin, FGF21 activated mTORC1 via MAPK rather than through the canonical PI3K/AKT pathway. Activation of mTORC1/S6K by FGF21 was surprising because this is thought to contribute to deleterious metabolic effects such as obesity and insulin resistance. Rather, mTORC1 mediated many of the beneficial actions of FGF21 in vitro, including UCP1 and FGF21 induction, increased adiponectin secretion, and enhanced glucose uptake without any adverse effects on insulin action. This study provides a global view of FGF21 signaling and suggests that mTORC1 may act to facilitate FGF21-mediated health benefits in vivo.

DNMT1 is essential for mammary and cancer stem cell maintenance and tumorigenesis.

Mammary stem/progenitor cells (MaSCs) maintain self-renewal of the mammary epithelium during puberty and pregnancy. DNA methylation provides a potential epigenetic mechanism for maintaining cellular memory during self-renewal. Although DNA methyltransferases (DNMTs) are dispensable for embryonic stem cell maintenance, their role in maintaining MaSCs and cancer stem cells (CSCs) in constantly replenishing mammary epithelium is unclear. Here we show that DNMT1 is indispensable for MaSC maintenance. Furthermore, we find that DNMT1 expression is elevated in mammary tumours, and mammary gland-specific DNMT1 deletion protects mice from mammary tumorigenesis by limiting the CSC pool. Through genome-scale methylation studies, we identify ISL1 as a direct DNMT1 target, hypermethylated and downregulated in mammary tumours and CSCs. DNMT inhibition or ISL1 expression in breast cancer cells limits CSC population. Altogether, our studies uncover an essential role for DNMT1 in MaSC and CSC maintenance and identify DNMT1-ISL1 axis as a potential therapeutic target for breast cancer treatment.

Direction pathway analysis of large-scale proteomics data reveals novel features of the insulin action pathway.

MOTIVATION: With the advancement of high-throughput techniques, large-scale profiling of biological systems with multiple experimental perturbations is becoming more prevalent. Pathway analysis incorporates prior biological knowledge to analyze genes/proteins in groups in a biological context. However, the hypotheses under investigation are often confined to a 1D space (i.e. up, down, either or mixed regulation). Here, we develop direction pathway analysis (DPA), which can be applied to test hypothesis in a high-dimensional space for identifying pathways that display distinct responses across multiple perturbations. RESULTS: Our DPA approach allows for the identification of pathways that display distinct responses across multiple perturbations. To demonstrate the utility and effectiveness, we evaluated DPA under various simulated scenarios and applied it to study insulin action in adipocytes. A major action of insulin in adipocytes is to regulate the movement of proteins from the interior to the cell surface membrane. Quantitative mass spectrometry-based proteomics was used to study this process on a large-scale. The combined dataset comprises four separate treatments. By applying DPA, we identified that several insulin responsive pathways in the plasma membrane trafficking are only partially dependent on the insulin-regulated kinase Akt. We subsequently validated our findings through targeted analysis of key proteins from these pathways using immunoblotting and live cell microscopy. Our results demonstrate that DPA can be applied to dissect pathway networks testing diverse hypotheses and integrating multiple experimental perturbations. AVAILABILITY AND IMPLEMENTATION: The R package 'directPA' is distributed from CRAN under GNU General Public License (GPL)-3 and can be downloaded from: http://cran.r-project.org/web/packages/directPA/index.html CONTACT: jean.yang@sydney.edu.au SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Dynamic adipocyte phosphoproteome reveals that Akt directly regulates mTORC2.

A major challenge of the post-genomics era is to define the connectivity of protein phosphorylation networks. Here, we quantitatively delineate the insulin signaling network in adipocytes by high-resolution mass spectrometry-based proteomics. These data reveal the complexity of intracellular protein phosphorylation. We identified 37,248 phosphorylation sites on 5,705 proteins in this single-cell type, with approximately 15% responding to insulin. We integrated these large-scale phosphoproteomics data using a machine learning approach to predict physiological substrates of several diverse insulin-regulated kinases. This led to the identification of an Akt substrate, SIN1, a core component of the mTORC2 complex. The phosphorylation of SIN1 by Akt was found to regulate mTORC2 activity in response to growth factors, revealing topological insights into the Akt/mTOR signaling network. The dynamic phosphoproteome described here contains numerous phosphorylation sites on proteins involved in diverse molecular functions and should serve as a useful functional resource for cell biologists.

Improving X!Tandem on peptide identification from mass spectrometry by self-boosted Percolator.

A critical component in mass spectrometry (MS)-based proteomics is an accurate protein identification procedure. Database search algorithms commonly generate a list of peptide-spectrum matches (PSMs). The validity of these PSMs is critical for downstream analysis since proteins that are present in the sample are inferred from those PSMs. A variety of postprocessing algorithms have been proposed to validate and filter PSMs. Among them, the most popular ones include a semi-supervised learning (SSL) approach known as Percolator and an empirical modeling approach known as PeptideProphet. However, they are predominantly designed for commercial database search algorithms, i.e., SEQUEST and MASCOT. Therefore, it is highly desirable to extend and optimize those PSM postprocessing algorithms for open source database search algorithms such as X!Tandem. In this paper, we propose a Self-boosted Percolator for postprocessing X!Tandem search results. We find that the SSL algorithm utilized by Percolator depends heavily on the initial ranking of PSMs. Starting with a poor PSM ranking list may cause Percolator to perform suboptimally. By implementing Percolator in a cascade learning manner, we can progressively improve the performance through multiple boost runs, enabling many more PSM identifications without sacrificing false discovery rate (FDR).

Re-fraction: a machine learning approach for deterministic identification of protein homologues and splice variants in large-scale MS-based proteomics.

A key step in the analysis of mass spectrometry (MS)-based proteomics data is the inference of proteins from identified peptide sequences. Here we describe Re-Fraction, a novel machine learning algorithm that enhances deterministic protein identification. Re-Fraction utilizes several protein physical properties to assign proteins to expected protein fractions that comprise large-scale MS-based proteomics data. This information is then used to appropriately assign peptides to specific proteins. This approach is sensitive, highly specific, and computationally efficient. We provide algorithms and source code for the current version of Re-Fraction, which accepts output tables from the MaxQuant environment. Nevertheless, the principles behind Re-Fraction can be applied to other protein identification pipelines where data are generated from samples fractionated at the protein level. We demonstrate the utility of this approach through reanalysis of data from a previously published study and generate lists of proteins deterministically identified by Re-Fraction that were previously only identified as members of a protein group. We find that this approach is particularly useful in resolving protein groups composed of splice variants and homologues, which are frequently expressed in a cell- or tissue-specific manner and may have important biological consequences.