Search

ObjectivesAcute kidney injury (AKI) is a frequent complication among critical ill patients with COVID-19, but the actual incidence is unknown as AKI-incidence varies from 25% to 89% in intensive care unit (ICU) populations. We aimed to describe the prevalence and risk factors of AKI in patients with COVID-19 admitted to ICU in Norway.DesignNation-wide observational study with data sampled from the Norwegian Intensive Care and Pandemic Registry (NIPaR) for the period between 10 March until 31 December 2020.SettingICU patients with COVID-19 in Norway. NIPaR collects data on intensive care stays covering more than 90% of Norwegian ICU and 98% of ICU stays.ParticipantsAdult patients with COVID-19 admitted to Norwegian ICU were included in the study. Patients with chronic kidney disease (CKD) were excluded in order to avoid bias from CKD on the incidence of AKI.Primary and secondary outcome measuresPrimary outcome was AKI at ICU admission as defined by renal Simplified Acute Physiology Score in NIPaR. Secondary outcome measures included survival at 30 and 90 days after admission to hospital.ResultsA total number of 361 patients with COVID-19 were included in the analysis. AKI was present in 32.0% of the patients at ICU admission. The risk for AKI at ICU admission was related to acute circulatory failure at admission to hospital. Survival for the study population at 30 and 90 days was 82.5% and 77.6%, respectively. Cancer was a predictor of 30-day mortality. Age, acute circulatory failure at hospital admission and AKI at ICU admission were predictors of both 30-day and 90-day mortality.ConclusionsA high number of patients with COVID-19 had AKI at ICU admission. The study indicates that AKI at ICU admission was related to acute circulatory failure at hospital admission. Age, acute circulatory failure at hospital admission and AKI at ICU admission were associated with mortality.

Using 11 proteomics datasets, mostly available through the PRIDE database, we assembled a reference expression map for 191 cancer cell lines and 246 clinical tumour samples, across 13 lineages. We found unique peptides identified only in tumour samples despite a much higher coverage in cell lines. These were mainly mapped to proteins related to regulation of signalling receptor activity. Correlations between baseline expression in cell lines and tumours were calculated. We found these to be highly similar across all samples with most similarity found within a given sample type. Integration of proteomics and transcriptomics data showed median correlation across cell lines to be 0.58 (range between 0.43 and 0.66). Additionally, in agreement with previous studies, variation in mRNA levels was often a poor predictor of changes in protein abundance. To our knowledge, this work constitutes the first meta-analysis focusing on cancer-related public proteomics datasets. We therefore also highlight shortcomings and limitations of such studies. All data is available through PRIDE dataset identifier PXD013455 and in Expression Atlas.

Mass spectrometry (MS)-based quantitative proteomics experiments typically assay a subset of up to 60% of the ≈20 000 human protein coding genes. Computational methods for imputing the missing values using RNA expression data usually allow only for imputations of proteins measured in at least some of the samples. In silico methods for comprehensively estimating abundances across all proteins are still missing. Here, a novel method is proposed using deep learning to extrapolate the observed protein expression values in label-free MS experiments to all proteins, leveraging gene functional annotations and RNA measurements as key predictive attributes. This method is tested on four datasets, including human cell lines and human and mouse tissues. This method predicts the protein expression values with average R 2 scores between 0.46 and 0.54, which is significantly better than predictions based on correlations using the RNA expression data alone. Moreover, it is demonstrated that the derived models can be "transferred" across experiments and species. For instance, the model derived from human tissues gave a R 2 = 0.51 when applied to mouse tissue data. It is concluded that protein abundances generated in label-free MS experiments can be computationally predicted using functional annotated attributes and can be used to highlight aberrant protein abundance values.

Lung cancer in East Asia is characterized by a high percentage of never-smokers, early onset and predominant EGFR mutations. To illuminate the molecular phenotype of this demographically distinct disease, we performed a deep comprehensive proteogenomic study on a prospectively collected cohort in Taiwan, representing early stage, predominantly female, non-smoking lung adenocarcinoma. Integrated genomic, proteomic, and phosphoproteomic analysis delineated the demographically distinct molecular attributes and hallmarks of tumor progression. Mutational signature analysis revealed age- and gender-related mutagenesis mechanisms, characterized by high prevalence of APOBEC mutational signature in younger females and over-representation of environmental carcinogen-like mutational signatures in older females. A proteomics-informed classification distinguished the clinical characteristics of early stage patients with EGFR mutations. Furthermore, integrated protein network analysis revealed the cellular remodeling underpinning clinical trajectories and nominated candidate biomarkers for patient stratification and therapeutic intervention. This multi-omic molecular architecture may help develop strategies for management of early stage never-smoker lung adenocarcinoma.

SummaryAssessing the extent to which genomic alterations compromise the integrity of the proteome is fundamental in identifying the mechanisms that shape cancer heterogeneity. We have used isobaric labelling and tribrid mass spectrometry to characterize the proteomic landscapes of 50 colorectal cancer cell lines and to decipher the relationships between genomic and proteomic variation. The robust quantification of 12,000 proteins and 27,000 phosphopeptides revealed how protein symbiosis translates to a co-variome which is subjected to a hierarchical order and exposes the collateral effects of somatic mutations on protein complexes. Targeted depletion of key chromatin modifiers confirmed the transmission of variation and the directionality as characteristics of protein interactions. Protein level variation was leveraged to build drug response predictive models towards a better understanding of pharmacoproteomic interactions in colorectal cancer. Overall, we provide a deep integrative view of the molecular structure underlying the variation of colorectal cancer cells.HighlightsThe cancer cell functional “co-variome” is a strong attribute of the proteome.Mutations can have a direct impact on protein levels of chromatin modifiers.Transmission of genomic variation is a characteristic of protein interactions.Pharmacoproteomic models are strong predictors of response to DNA damaging agents.AbbreviationsCOREADColorectal AdenocarcinomaIMACImmobilized Metal ion Affinity ChromatographyROCReceiver Operating CharacteristicAUCArea Under the CurveWGCNAWeighted Correlation Network AnalysisCNACopy Number AlterationSOMSelf-Organizing MapQTLQuantitative Trait LociMSIMicrosatellite InstabilityCPSColorectal Proteomic Subtypes

Summary Assessing the impact of genomic alterations on protein networks is fundamental in identifying the mechanisms that shape cancer heterogeneity. We have used isobaric labeling to characterize the proteomic landscapes of 50 colorectal cancer cell lines and to decipher the functional consequences of somatic genomic variants. The robust quantification of over 9,000 proteins and 11,000 phosphopeptides on average enabled the de novo construction of a functional protein correlation network, which ultimately exposed the collateral effects of mutations on protein complexes. CRISPR-cas9 deletion of key chromatin modifiers confirmed that the consequences of genomic alterations can propagate through protein interactions in a transcript-independent manner. Lastly, we leveraged the quantified proteome to perform unsupervised classification of the cell lines and to build predictive models of drug response in colorectal cancer. Overall, we provide a deep integrative view of the functional network and the molecular structure underlying the heterogeneity of colorectal cancer cells., Graphical Abstract, Highlights • The cancer cell “co-variome” recapitulates functional protein associations • Loss-of-function mutations can impact protein levels beyond mRNA regulation • The consequences of genomic alterations can propagate through protein interactions • We provide insight into the molecular organization of colorectal cancer cells, Roumeliotis et al. use in-depth proteomics to assess the impact of genomic alterations on protein networks in colorectal cancer cell lines. Cell-line-specific network signatures are inferred de novo by protein quantification profiles and ultimately expose the collateral and transcript-independent effects of detrimental mutations on protein complexes.

The analysis of polygenetic characteristics for mapping quantitative trait loci (QTL) remains an important challenge. QTL analysis requires two or more strains of organisms that differ substantially in the (poly-)genetic trait of interest, resulting in a heterozygous offspring. The offspring with the trait of interest is selected and subsequently screened for molecular markers such as single-nucleotide polymorphisms (SNPs) with next-generation sequencing. Gene mapping relies on the co-segregation between genes and/or markers. Genes and/or markers that are linked to a QTL influencing the trait will segregate more frequently with this locus. For each identified marker, observed mismatch frequencies between the reads of the offspring and the parental reference strains can be modeled by a multinomial distribution with the probabilities depending on the state of an underlying, unobserved Markov process. The states indicate whether the SNP is located in a (vicinity of a) QTL or not. Consequently, genomic loci associated with the QTL can be discovered by analyzing hidden states along the genome. The aforementioned hidden Markov model assumes that the identified SNPs are equally distributed along the chromosome and does not take the distance between neighboring SNPs into account. The distance between the neighboring SNPs could influence the chance of co-segregation between genes and markers. To address this issue, we propose a nonhomogeneous hidden Markov model with a transition matrix that depends on a set of distance-varying observed covariates. The application of the model is illustrated on the data from a study of ethanol tolerance in yeast.

To extract a genuine peptide signal from a mass spectrum, an observed series of peaks at a particular mass can be compared with the isotope distribution expected for a peptide of that mass. To decide whether the observed series of peaks is similar to the isotope distribution, a similarity measure is needed. In this short communication, we investigate whether the Mahalanobis distance could be an alternative measure for the commonly employed Pearson's chi(2) statistic. We evaluate the performance of the two measures by using a controlled MALDI-TOF experiment. The results indicate that Pearson's chi(2) statistic has better discriminatory performance than the Mahalanobis distance and is a more robust measure.

Shotgun proteomics is a powerful technology to study the protein population of a biological system. This approach employs tandem mass spectrometry for amino acid sequencing. Fragmented ion masses can be used in correlative database-searching, like SEQUEST or Mascot, to identify peptides. The database-search method depends upon a score function that evaluates matches between the predicted ions and the ions observed in the tandem mass spectrum. Principally, peptide identification based on tandem MS and database-search algorithms does not take into account information about isotope distributions of the precursor ions. To determine the effectiveness of these search algorithms in terms of their ability to distinguish between correct and incorrect peptide assignments, we propose an additional metric that quantifies the similarity between the theoretical isotopic distribution for the precursor ions selected for tandem MS and the experimental mass spectra by using Pearson's χ2 statistic. The observed association between Pearson's χ2 statistic and the score function indicates that good scores can be obtained for molecules which exhibit atypical isotope profiles, while low scores can be obtained for fragment spectra which have a clear peptide-like isotope pattern. These results demonstrate that Pearson's χ2 statistic can be used in conjunction with the score of database-search algorithms to increase the sensitivity and specificity of peptide identification. Biological significance In this manuscript, we present a workflow that provides a new perspective on the quality of peptide-to-spectrum matches (PSM) employed in database-searching strategies for peptide identification. Additional views on a dataset can facilitate a more hypothesis-driven interpretation of the mass spectrometry signals. The similarity metric on the PSM scores contemplates the isotopic profile and results in a measure that conveys a degree of biomolecular similarity observed from the precursor of the selected tandem MS spectra. A close agreement between the PSM score and the similarity metric will result in a higher confidence for the identification of the selected precursor ion.

To extract a genuine peptide signal from a mass spectrum, an observed series of peaks at a particular mass can be compared with the isotope distribution expected for a peptide of that mass. To decide whether the observed series of peaks is similar to the isotope distribution, a similarity measure is needed. In this short communication, we investigate whether the Mahalanobis distance could be an alternative measure for the commonly employed Pearson's χ(2) statistic. We evaluate the performance of the two measures by using a controlled MALDI-TOF experiment. The results indicate that Pearson's χ(2) statistic has better discriminatory performance than the Mahalanobis distance and is a more robust measure.