Unsupervised Analysis Research Articles

Clinicopathological and epigenetic differences between primary neuroendocrine tumors and neuroendocrine metastases in the ovary.

Currently, the available literature provides insufficient support to differentiate between primary ovarian neuroendocrine tumors (PON) and neuroendocrine ovarian metastases (NOM) in patients. For this reason, patients with a well-differentiated ovarian neuroendocrine tumor (NET) were identified through electronic patient records and a nationwide search between 1991 and 2023. Clinical characteristics were collected from electronic patient files. This resulted in the inclusion of 71 patients with NOM and 17 patients with PON. Histologic material was stained for Ki67, SSTR2a, CDX2, PAX8, TTF1, SATB2, ISLET1, OTP, PDX1, and ARX. DNA methylation analysis was performed on a subset of cases. All PON were unilateral and nine were found within a teratoma (PON-T+). A total of 78% of NOM were bilateral, and none were associated with a teratoma. PON without teratomous components (PON-T-) displayed a similar insular growth pattern and immunohistochemistry as NOM (p > 0.05). When compared with PON-T+, PON-T- more frequently displayed ISLET1 positivity and were larger, and patients were older at diagnosis (p < 0.05). Unsupervised analysis of DNA methylation profiles from tumors of ovarian (n = 16), pancreatic (n = 22), ileal (n = 10), and rectal (n = 7) origin revealed that four of five PON-T- clustered together with NOM and ileal NET, whereas four of five PON-T+ grouped with rectum NET. In conclusion, unilateral ovarian NET within a teratoma should be treated as a PON. Ovarian NET localizations without teratomous components have a molecular profile analogous to midgut NET metastases. For these patients, a thorough review of imaging should be performed to identify a possible undetected midgut NET and a corresponding follow-up strategy may be recommended.

Unsupervised Histopathological Sub-Image Analysis for Breast Cancer Diagnosis Using Variational Autoencoders, Clustering, and Supervised Learning

This paper presents an integrated approach to breast cancer diagnosis that combines unsupervised and supervised learning techniques. The method involves using a pre-trained VGG19 model to process sub-images from the BreaKHis dataset, divided into nine parts for comprehensive analysis. This will be followed by a complete description of the architecture and workings of the variational Autoencoder (VAE) used for unsupervised Learning. The encoder network maps the input features to lower dimensions, capturing the most essential information. VAE learns a compressed representation of sub-images, facilitating a more profound understanding of underlying patterns and structures. For this reason, we then employ k-means clustering on the encoded representation to find naturally occurring clusters in our data set comprising a histopathological image. Every single sub-image is later fed into the VGG19-SVM model for classification purposes. During magnification at 100x, this model has attained a fantastic accuracy rate of 98.56%. Combining unsupervised analysis with VAE/k-means clustering and supervised classification with VGG19/SVM can integrate information from both methods, thereby improving the accuracy and robustness of such a task as sub-image classification in breast cancer histopathology.

Discovery of PANoptosis-related signatures correlates with immune cell infiltration in psoriasis.

Psoriasis is an inflammatory skin disease that relapses frequently. Keratinocyte apoptosis dysregulation plays a crucial role in the pathological mechanisms of psoriasis. PANoptosis is a process with intermolecular interaction among pyroptosis, apoptosis, and necroptosis. The mechanism of PANoptosis in the occurrence and development of psoriasis is still unclear. Here we present a novel approach by identifying PANoptosis-related signatures (PANoptosis-sig) from skin tissue of psoriasis patients and healthy controls on transcriptional and protein levels. Five PANoptosis-sig (TYMP, S100A8, S100A9, NAMPT, LCN2) were identified. Enrichment analysis showed they were mainly enriched in response to leukocyte aggregation, leukocyte migration, chronic inflammatory response and IL-17 signaling pathway. Single cell transcriptome analysis showed TYMP and NAMPT were expressed in almost all cell populations, while LCN2, S100A8 and S100A9 were significantly highly expressed in keratinocyte. We then constructed predictive and diagnostic models with the PANoptosis-sig and evaluated their performance. Finally, unsupervised consensus clustering analysis was conducted to ascertain psoriasis molecular subtypes by the PANoptosis-sig. The psoriasis cohort was divided into two distinct subtypes. Immune landscape showed that the stromal score of cluster 1 was significantly higher than cluster 2, while the immune and estimate scores of cluster 2 were expressively higher than cluster 1. Cluster 1 exhibited high expression of Plasma cells, Tregs and Mast cells resting, while cluster 2 showed high expression of T cells, Macrophages M1, Dendritic cells activated, and Neutrophils in immune infiltration analysis. And cluster 2 was more sensitive to immune checkpoints. In conclusion, our findings revealed potential biomarkers and therapeutic targets for the prevention, diagnosis, and treatment of psoriasis, enhancing our understanding of the molecular mechanisms underlying PANoptosis.

AI-based cluster analysis enables outcomes prediction among patients with increased LVM

Abstract Background The traditional classification of left ventricular hypertrophy (LVH) solely based on left ventricular geometry disregards associated comorbidities and clinical characteristics. We aimed to identify unique clinical phenotypes of LVH using unsupervised cluster analysis and explore their association with clinical outcomes. Methods From UK Biobank (UKBB) participants with available left ventricular mass (LVM) estimations, we identified those meeting CMR-based criteria for increased LVM: LVM index ≥ 72 g/m2 for males and LVM index ≥ 55 g/m2 for females. Baseline demographic, clinical, and laboratory data were collected. Using Ward's method, patients were clustered based on 27 variables. The primary outcome was a composite of all-cause mortality with heart failure (HF) admissions, ventricular arrhythmia, and atrial fibrillation (AF). Cox proportional hazard model and Kaplan-Meier survival analysis were applied. Results Among CMR-assessed UKBB participants, 4,255 individuals exhibited increased LVM, with an average age of 64 ± 7 years; 2,447 (58%) were females. Cluster analysis identified four distinct subgroups. Over a median 5-year follow-up (IQR: 4-6), 100 patients (2%) died, 118 (2.8%) were hospitalized due to HF, 29 (0.7%) were hospitalized due to VA, and 208 (5%) were hospitalized due to AF. Univariate Cox analysis revealed that compared to the 1st cluster (n=1,578), patients in the 2nd (n=1,296), 3rd (n=824) and 4th (n=557) clusters had 60%, 104%, and 164% increased risk of a major event, respectively (95% CI 1.2-2.16, p&amp;lt;.001 for cluster 2; 95% CI 1.49-2.78, p&amp;lt;.001 for cluster 3; 95% CI 1.92-3.62, p&amp;lt;.001 for cluster 4). Each cluster manifested different phenotypes: cluster 2 comprised mainly overweight females, with the highest prevalence of chronic lung disease; cluster 3, predominantly composed of males, had the highest burden of comorbidities and cardiovascular risk factors; and cluster 4, mostly males, presented with the most abnormal cardiac measures, including left ventricular end-diastolic volume (LVEDV), left ventricular end-systolic volume (LVESV), and left ventricular ejection fraction (LVEF). Conclusions Unsupervised cluster analysis identified four subgroups among patients with increased LVM. These subgroups bear notable relevance to overall mortality risk and clinical outcomes. This phenotypic classification holds promise in offering valuable insights regarding clinical course and outcomes in LVH patients.

Prognostic value of myocardial redox signalling: a machine learning multi-omics approach

Abstract Background Oxidative stress has been proposed to be implicated in cardiovascular pathology. Studies have linked myocardial oxidative stress with the development of atrial fibrillation, but its role in the prediction of long-term outcomes and mortality is yet to be elucidated. Purpose To apply discovery network transcriptomics and genomics to atrial tissue (AT) obtained from patients undergoing cardiac surgery, to phenotype myocardial redox-state and identify potential therapeutic targets. Methods We included 1,032 patients undergoing cardiac surgery. Arm 1 included 258 patients in whom atrial tissue (right atrial appendage) was used for ex-vivo quantification of NOX-derived superoxide by lucigenin-enhanced chemiluminescence. Arm 2 included 414 patients in whom exploratory, unsupervised gene network analysis in human atrial tissue RNA sequencing data was performed to identify atrial tissue redox-sensitive signatures (blue signature). Arm 3 included 881 patients genotyped with the genome-wide SNP array UK Biobank Axiom Array and used for machine learning construction of GenRedOx, a SNP model built to predict the blue transcriptomic signature. Associations with future incidence of major adverse myocardial events (MAME: cardiovascular death and/or heart failure) was assessed across all arms using cox regression models (adjusted for age, sex, hypertension, dyslipidaemia, diabetes mellitus, body mass index, and plasma TNFa). Results Over a median follow-up of 5.27 years [IQR: 4.29-6.83] high atrial tissue (AT) NOX-derived superoxide was independently associated with MAME risk (Adj. HR[95%CI]: 11.7 [2.18 , 62.82], p=0.004, a). Unsupervised transcriptomic analysis in Arm 2 revealed 15 coexpressed gene ‘modules’ or 'signatures' (b). The blue signature was significantly associated with AT superoxide production with "immune response-regulating and activating signalling" being amongst the top enriched pathways. Patients with high eigengene values for the blue signature showed a 5.5-fold higher adjusted risk of MAME (c). In Arm 3, extreme gradient boosting of SNP data, trained using the blue transcriptomic signature as ground truth, revealed GenRedOx (d), a genomic artificial intelligence score, which was found to be prognostic of MAME in the validation stage (e). Homozygosity for the minor allele of SNP rs723897 in the Chemokine-Like Factor Super Family 7 (CMTM7) gene had the highest gain in the model. Conclusion We present for the first time two novel human myocardial transcriptomic and genomic signatures that reflect changes in redox state and identify long-term cardiovascular risk. Targeting pathways in the myocardium related with immune response signalling may lead to the development of novel therapeutics in cardiovascular disease.

Proteome profiling in patients with peripartum cardiomyopathy according to ethnicity a biomarker study on the EORP cohort

Abstract Background Peripartum cardiomyopathy (PPCM) is a pregnancy-associated form of heart failure, which occurs in previously healthy women towards the end of pregnancy and in the first months postpartum. Previous reports showed a varying prevalence of PPCM across different regions and potentially a higher incidence in patients with black African ethnicity. The correct and timely diagnosis remains challenging and there is a need for novel disease-specific biomarkers. Accuracy in diagnosing PPCM can be improved by using serum-protein biomarkers such as QSOX1, adiponectin and ITIH3 in addition to NT-proBNP. However, it still needs to be ascertained whether there are ethnic differences in proteomic profiles amongst patients with PPCM with different ethnicities. Aims To delineate possible differences in serum protein biomarker profiles across different ethnic groups within the EURObservation Research Programme (EORP) multi-national cohort of women with PPCM. Methods Demographic and clinical data, as well as serum samples were collected from 84 patients with PPCM from seven EORP participating countries. Serum proteomic profiling was conducted using DIA-based label-free quantitative (LFQ) LC-MS at the time of diagnosis from depleted serum samples. Mass spectrometry data were analyzed by Spectronaut v15 using a study-specific spectral library. From the 84 patients in the EORP PPCM registry on whom we performed proteomic analyses, 82 patients self-declared their ethnicity. These 82 patients were assessed for any ethnical differences in their proteomic expression profile using unsupervised principal component analysis (PCA). Results In this cohort of 82 women with PPCM, the mean age was 30.5±6.7 years (Table 1). Three quarters of this cohort had no hypertension during pregnancy and only 13.8% had a prior diagnosis of PPCM. The mean left ventricular ejection fraction (LVEF) was 35% at time of first diagnosis, with no difference between the ethnic groups. However, women of Middle Eastern decent had more severe LV dilatation (LVEDD: 64.0 (62.0-65.0) mm). Three hundred and twenty-nine (329) proteins were identified in the serum samples. As depicted in Figure 1, PCA did not yield a distinct separation between the different ethnic groups. Principal component 1 (PC1) accounted for 15.2% of the total variability in protein expression among ethnic groups, PC2 explained a comparatively lower percentage at 12%. Conclusion The protein biomarkers identified in PPCM patients showed no discernible racial differences, demonstrating consistent performance across all ethnic groups.

Latent class analysis reveals a novel phenotype of heart failure that spans the left ventricular ejection fraction range

Abstract Background It is increasingly evident that heart failure (HF) represents a heterogenous syndrome rather than a single clinical entity. The elusiveness of effective therapies for heart failure with preserved ejection fraction (HFpEF) has led to a consensus that our current, unidimensional construct of HF using only left ventricular ejection fraction (LVEF) does not adequately capture the multiplicity of the disease states. The ability to harmonize echocardiographic data from multiple sources has enabled use of unsupervised cluster-based analysis to identify multidimensional cardiac structural phenotypes. Purpose To identify cardiac structural phenotypes of HF across the LVEF spectrum. Methods We harmonized echocardiographic data from eight randomized clinical HF trials (ESCAPE, EXACT-HF, FIGHT, I-PRESERVE, NEAT-HFpEF, RELAX, STICH and TOPCAT), which included both HFrEF and HFpEF populations (total n=4223 echocardiograms). These studies collectively spanned the years 2000 through 2015 and represent a balanced LVEF distribution. Latent class analysis (LCA) was performed iteratively to identify predominant structural phenotypes using LVEF, RV function, LV end diastolic volume, LV mass index, eccentricity, mitral valve deceleration time, and lateral E/e’ ratio. Results LCA identified 4 echocardiographic phenotypes. Three phenotypes skewed toward either high or low EF, but one phenotype demonstrated a relatively even distribution of patients with HFrEF, heart failure with mildly reduced ejection fraction (HFmrEF) and HFpEF (36.0%, 21.4% and 42.7%, respectively). Patients in this phenotype had preserved RV function and left ventricular hypertrophy (97%), of which 56.8% had concentric and 43.2% eccentric LVH. E/e’ was normal in 63% of subjects. A vast majority (91%) of patients had recent-onset HF (diagnosed within the past year). There were no remarkable distinguishing comorbidities. Notable differences in biomarker profile included higher levels of troponin I, cyclic GMP and lower levels of neurohormones (aldosterone, endothelin-1, dopamine, and norepinephrine) than the other phenotypes. Conclusions This is the first cluster-based analysis to identify a quantitative structural HF phenotype of HF that is not distinguished by LVEF. While the underlying pathophysiology is unclear, most patients have recent onset HF and biomarker profiles with relatively low neurohormone activity. These features may reflect a distinct pathophysiology that could indicate a specific, targeted intervention.

Clinical phenotypes in relation to outcomes in heart failure patients with cardiac resynchronization therapy and defibrillators (CRT-D): an unsupervised cluster analysis

Abstract Background Patients with heart failure undergoing CRT are an heterogenous and complex population. Selecting patients before implantation is essential to obtain a favorable response to CRT-D. Data-driven cluster analysis may be an approach to identify different patients’ phenotypic categories. Purpose To identify different clusters of patients with CRT-D who share similar clinical phenotypes, and to evaluate the associations between clusters and clinical outcomes, using cluster analysis. Methods Three agglomerative hierarchical cluster analysis were performed in CRT-D patients seen between 2010 and 2019 in French hospitals. Associations between clusters and death at one year and death during whole follow-up (FU) were evaluated using Cox regression analyses. Results The study included 23,029 CRT-D patients with no prior history of VT/VF/cardiac arrest (mean age 67.7 ± 9.9 years; 78.8 % male), who were analyzed in relation to 3 ways, as follows: the first group was a 50% random sample of all patients (n = 11,514), the second group included patients dead at 1 year (n = 1,604) and the third group included those alive at 3 years FU (n = 14,228). A cluster analysis was performed on each group. Four clusters were identified in the first group: Cluster 1 identified young patients with dilated cardiomyopathy and low prevalence of coronary artery disease (CAD), cardiovascular (CV) risk factors and comorbidities (chronic kidney disease, lung and liver disease) (low-risk phenotype); Cluster 2 was composed of male patients with CAD but low burden of CV risk factors and comorbidities (CAD phenotype); Cluster 3 included CRT recipients with several CV risk factors but low prevalence of comorbidities (CV risk factors phenotype); Cluster 4 identified old patients with high prevalence of CAD, atrial fibrillation, CV risk factors and comorbidities (clinical complex phenotype). These phenotypes were maintained in the other two groups (dead at 1 year and alive at 3 years FU), with the highest consistency for low-risk and high-risk phenotypes(Cluster 1 and 4 respectively). Compared with Cluster 1, Clusters 2, 3 and 4 were independently associated with an increased risk of all-cause death at 1-year FU and during the whole FU (Cluster 2: HR 1.21, 95% CI 1.08 - 1.36; Cluster 3: HR 1.15, 95%CI 1.04 - 1.26; and Cluster 4: HR 1.79, 95%CI 1.65 - 1.96). There were high prevalences of CAD, atrial fibrillation, mitral regurgitation, and comorbidities (which are all predictors of CRT-nonresponse) not only in Cluster 4 at baseline, but also among patients who died at 1-year FU in all 4 clusters. Conclusion Cluster analysis identified four statistically driven groups of CRT recipients, with specific clinical phenotypes and associated with different risks for all-cause death. Identifying those patients at higher risk of adverse events might help in the selection of patients with heart failure undergoing CRT-D implantation, especially in relation to CRT-nonresponse.Risk of all-cause death in 4 clusters

Modeling a biofluid-derived extracellular vesicle surface signature to differentiate pediatric idiopathic nephrotic syndrome clinical subgroups

Idiopathic Nephrotic Syndrome (INS) is a common childhood glomerular disease requiring intense immunosuppressive drug treatments. Prediction of treatment response and the occurrence of relapses remains challenging. Biofluid-derived extracellular vesicles (EVs) may serve as novel liquid biopsies for INS classification and monitoring. Our cohort was composed of 105 INS children at different clinical time points (onset, relapse, and persistent proteinuria, remission, respectively), and 19 healthy controls. The expression of 37 surface EV surface markers was evaluated by flow cytometry in serum (n = 83) and urine (n = 74) from INS children (mean age = 10.1, 58% males) at different time points. Urine EVs (n = 7) and serum EVs (n = 11) from age-matched healthy children (mean age = 7.8, 94% males) were also analyzed. Tetraspanin expression in urine EVs was enhanced during active disease phase in respect to the remission group and positively correlates with proteinuria levels. Unsupervised clustering analysis identified an INS signature of 8 markers related to immunity and angiogenesis/adhesion processes. The CD41b, CD29, and CD105 showed the best diagnostic scores separating the INS active phase from the healthy condition. Interestingly, combining urinary and serum EV markers from the same patient improved the precision of clinical staging separation. Three urinary biomarkers (CD19, CD44, and CD8) were able to classify INS based on steroid sensitivity. Biofluid EVs offer a non-invasive tool for INS clinical subclassification and “personalized” interventions.

Machine learning-based new classification for immune infiltration of gliomas.

Glioma is a highly heterogeneous and poorly immunogenic malignant tumor, with limited efficacy of immunotherapy. The characteristics of the immunosuppressive tumor microenvironment (TME) are one of the important factors hindering the effectiveness of immunotherapy. Therefore, this study aims to reveal the immune microenvironment (IME) characteristics of glioma and predict different immune subtypes using machine learning methods, providing guidance for immune therapy in glioma. We first performed unsupervised cluster analysis on the genes and arrays of 693 gliomas in CGGA database and 702 gliomas in TCGA database. Then establish and verify the classification model through Machine Learning (ML). Then, use DAVID to perform functional enrichment analysis for different immune subtypes. Next step, analyze the immune cell distribution, stemness maintenance, mesenchymal phenotype, neuronal phenotype, tumorigenic cytokines, molecular and clinical characteristics of different immune subtypes of gliomas. Firstly, we divide the IME of gliomas in the CGGA database into four different subtypes, namely IM1, IM2, IM3, and IM4; similarly, the IME of gliomas in the TCGA database can also be divided into four different subtypes (IMA, IMB, IMC, and IMD). Next, based on ML, we developed a highly reliable model for predicting different immune subtypes of glioma. Then, we found that Monocytic lineage, Myeloid dendritic cells, NK cells and CD8 T cells had the highest enrichment in the IM1/IMD subtypes. Cytotoxic lymphocytes were highest expressed in the IM4/IMA subtypes. Next step, Enrichment analysis revealed that the IM1-IMD subtypes were mainly closely related to the production and secretion of IL-8 and TNF signaling pathway. The IM2-IMB subtypes were strongly associated with leukocyte activation and NK cell mediated cytotoxicity. The IM3-IMC subtypes were closely related to mitotic nuclear division and mitotic cell cycle process. The IM4-IMA subtypes were strongly associated with Central Nervous System (CNS) development and striated muscle tissue development. Afterwards, Single sample gene set enrichment analysis (ssGSEA) showed that stemness maintenance phenotypes were mainly enriched in the IM4/IMA subtypes; Neuronal phenotypes were closely associated with the IM2/IMB subtypes; and mesenchymal phenotypes and tumorigenic cytokines were highly correlated with the IM2 /IMB subtypes. Finally, we found that compared with patients in the IM2/IMB and IM4/IMA subtypes, the IM1/IMD and IM3/IMC subtypes have the highest proportion of GBM patients, the shortest average overall survival of patients and the lowest proportion of patients with IDH mutation and 1p36/19q13 co-deletion. We developed a highly reliable model for predicting different immune subtypes of glioma by ML. Then, we comprehensively analyzed the immune infiltration, molecular and clinical features of different immune subtypes of gliomas and defined gliomas into four subtypes: immunogenic subtype, adaptive immune resistance subtype, mesenchymal subtype, and immune tolerance subtype, which represent different TMEs and different stages of tumor development.

Construction of mitochondrial quality regulation genes-related prognostic model based on bulk-RNA-seq analysis in multiple myeloma.

Mitochondrial quality regulation plays an important role in affecting the treatment sensitivity of multiple myeloma (MM). We aimed to develop a mitochondrial quality regulation genes (MQRGs)-related prognostic model for MM patients. The Genomic Data Commons-MM of bulk RNA-seq, mutation, and single-cell RNA-seq (scRNA-seq) dataset were downloaded, and the MQRGs gene set was collected previous study. "maftools" and CIBERSORT were used for mutation and immune-infiltration analysis. Subsequently, the "ConsensusClusterPlus" was used to perform the unsupervised clustering analysis, "survminer" and "ssGSEA" R package was used for the Kaplan-Meier survival and enrichment analysis, "limma" R, univariate and Least Absolute Shrinkage and Selection Operator Cox were used for RiskScore model. The "timeROC" R package was used for Receiver Operating Characteristic Curve analysis. Finally, the "Seurat" R package was used for scRNA-seq analysis. These MQRGs are mainly located on chromosome-1,2,3,7, and 22 and had significant expression differences among age, gender, and stage groups, in which PPARGC1A and PPARG are the high mutation genes. Most MQRGs expression are closely associated with the plasma cells infiltration and can divide the patients into 2 different prognostic clusters (C1, C2). Then, 8 risk models were screened from 60 DEGs for RiskScore, which is an independent prognostic factor and effectively divided the patients into high and low risk groups with significant difference of immune checkpoint expression. Nomogram containing RiskScore can accurately predict patient prognosis, and a series of specific transcription factor PRDM1 and IRF1 were identified. We described the based molecular features and developed a high effective MQRGs-related prognostic model in MM.

Archetypal Analysis of Kidney Allograft Biopsies Using Next-generation Sequencing Technology.

In kidney transplantation, molecular diagnostics may be a valuable approach to improve the precision of the diagnosis. Using next-generation sequencing (NGS), we aimed to identify clinically relevant archetypes. We conducted an Illumina bulk RNA sequencing on 770 kidney biopsies (540 kidney recipients) collected between 2006 and 2021 from 11 European centers. Differentially expressed genes were determined for 11 Banff lesions. An ElasticNet model was used for feature selection, and 4 machine learning classifiers were trained to predict the probability of presence of the lesions. NGS-based classifiers were used in an unsupervised archetypal analysis to different archetypes. The association of the archetypes with allograft survival was assessed using the iBox risk prediction score. The ElasticNet feature selection reduced the number of the genes from a range of 859-10 830 to a range of 52-867 genes. NGS-based classifiers demonstrated robust performances (precision-recall area under the curves 0.708-0.980) in predicting the Banff lesions. Archetypal analysis revealed 8 distinct phenotypes, each characterized by distinct clinical, immunological, and histological features. Although the archetypes confirmed the well-defined Banff rejection phenotypes for T cell-mediated rejection and antibody-mediated rejection, equivocal histologic antibody-mediated rejection, and borderline diagnoses were reclassified into different archetypes based on their molecular signatures. The 8 NGS-based archetypes displayed distinct allograft survival profiles with incremental graft loss rates between archetypes, ranging from 90% to 56% rates 7 y after evaluation (P < 0.0001). Using molecular phenotyping, 8 archetypes were identified. These NGS-based archetypes might improve disease characterization, reclassify ambiguous Banff diagnoses, and enable patient-specific risk stratification.

Proteomic analysis of post-COVID condition: Insights from plasma and pellet blood fractions

BackgroundPersistent symptoms extending beyond the acute phase of SARS-CoV-2 infection, known as Post-COVID condition (PCC), continue to impact many individuals years after the COVID-19 pandemic began. This highlights an urgent need for a deeper understanding and effective treatments. While significant progress has been made in understanding the acute phase of COVID-19 through omics-based approaches, the proteomic alterations linked to the long-term effects of the infection remain underexplored. This study aims to investigate these proteomic changes and develop a method for stratifying disease severity. MethodsUsing Sequential Window Acquisition of All Theoretical Fragment Ion Mass Spectra (SWATH-MS) technology, we performed comprehensive proteomic profiling of blood samples from 65 PCC patients. Both plasma and pellet (cellular components) fractions were analyzed to capture a wide array of proteomic changes associated with PCC. ResultsProteomic profiling revealed distinct differences between symptomatic and asymptomatic PCC patients. In the plasma fraction, symptomatic patients exhibited significant upregulation of proteins involved in coagulation, immune response, oxidative stress, and various metabolic processes, while certain immunoglobulins and proteins involved in cellular stress responses were downregulated. In the pellet fraction, symptomatic patients showed upregulation of proteins related to immune response, coagulation, oxidative stress, and metabolic enzymes, with downregulation observed in components of the complement system, glycolysis enzymes, and cytoskeletal proteins. A key outcome was the development of a novel severity scale based on the concentration of identified proteins, which correlated strongly with the clinical symptoms of PCC. This scale, derived from unsupervised clustering analysis, provides precise quantification of PCC severity, enabling effective patient stratification. ConclusionsThe identified proteomic alterations offer valuable insights into the molecular mechanisms underlying PCC, highlighting potential biomarkers and therapeutic targets. This research supports the development of tailored clinical interventions to alleviate persistent symptoms, ultimately enhancing patient outcomes and quality of life. The quantifiable measure of disease severity aids clinicians in understanding the condition in individual patients, facilitating personalized treatment plans and accurate monitoring of disease progression and response to therapy.

Immunological sub-phenotypes and response to convalescent plasma in COVID-19 induced ARDS: a secondary analysis of the CONFIDENT trial

BackgroundConvalescent plasma (CP) reduced the mortality in COVID-19 induced ARDS (C-ARDS) patients treated in the CONFIDENT trial. As patients are immunologically heterogeneous, we hypothesized that clusters may differ in their treatment responses to CP.MethodsWe measured 20 cytokines, chemokines and cell adhesion markers using a multiplex technique at the time of inclusion in the CONFIDENT trial in patients of centers having accepted to participate in this secondary study. We performed descriptive statistics, unsupervised hierarchical cluster analysis, and examined the association between the clusters and CP effect on day-28 mortality.ResultsOf the 475 patients included in CONFIDENT, 391 (82%) were sampled, and 196/391 (50.1%) had been assigned to CP. We identified four sub-phenotypes representing 89 (22.8%), 178 (45.5%), 38 (9.7%), and 86 (22.0%) patients. The most contributing biomarkers in the principal component analysis were IL-1β, IL-12p70, IL-6, IFN-α, IL-17A, IFN-γ, IL-13, TFN-α, total IgG, and CXCL10. Sub-phenotype-1 displayed a lower immune response, sub-phenotype-2 a higher adaptive response, sub-phenotype-3 the highest innate antiviral, pro and anti-inflammatory response, and adhesion molecule activation, and sub-phenotype-4 a higher pro and anti-inflammatory response, migration protein and adhesion molecule activation. Sub-phenotype-2 and sub-phenotype-4 had higher severity at the time of inclusion. The effect of CP treatment on mortality appeared higher than standard care in each sub-phenotype, without heterogeneity between sub-phenotypes (p = 0.97).ConclusionIn patients with C-ARDS, we identified 4 sub-phenotypes based on their immune response. These sub-phenotypes were associated with different clinical profiles. The response to CP was similar across the 4 sub-phenotypes.Trial registration: Ethics Committee of the University Hospital of Liège CE 2020/239. Clinicaltrials.gov NCT04558476. Registered 2020-09-11, https://www.clinicaltrials.gov/study/NCT04558476.

Circulating B Cells Display Differential Immune Regulatory Molecule Expression in Granulomatosis with Polyangiitis.

Granulomatosis with polyangiitis (GPA) is a B cell-mediated, relapsing, autoimmune disease. There is a need for novel therapeutic approaches and relapse markers to achieve durable remission. B cells express immune regulatory molecules that modulate their activation and maintain tolerance. While recent studies show dysregulation of these molecules in other autoimmune diseases, data on their expression in GPA are limited. This study aimed to map the expression of surface immune regulatory molecules on circulating B cell subsets in GPA and correlate their expression with clinical parameters. Immune regulatory molecule expression on circulating B cell subsets was comprehensively examined in active GPA (n=16), GPA in remission (n=16), and healthy controls (HCs, n=16) cross-sectionally using a 35-color B cell-specific spectral flow cytometry panel. Our supervised and unsupervised in-depth analysis revealed differential expression of inhibitory and stimulatory immune molecules on distinct B cell populations in GPA, with the most notable differences observed in active GPA. These differences include the upregulation of FcγRIIB on non-mature B cells, downregulation of CD21 and upregulation of CD86 on antigen-experienced B cells, and elevated CD22 expression on various populations. Additionally, we found a strong association between FcγRIIB, BTLA, and CD21 expression on specific B cell populations and disease activity in GPA. Together, these findings provide novel insights into the immune regulatory molecule expression profile of B cells in GPA, and could potentially form the foundation for new therapeutic approaches and disease monitoring markers.

Peripheral immune profiling of soft tissue sarcoma: perspectives for disease monitoring.

Studying the tumor microenvironment and surrounding lymph nodes is the main focus of current immunological research on soft tissue sarcomas (STS). However, due to the restricted opportunity to examine tumor samples, alternative approaches are required to evaluate immune responses in non-surgical patients. Therefore, the purpose of this study was to evaluate the peripheral immune profile of STS patients, characterize patients accordingly and explore the impact of peripheral immunotypes on patient survival. Blood samples were collected from 55 STS patients and age-matched healthy donors (HD) controls. Deep immunophenotyping and gene expression analysis of whole blood was analyzed using multiparametric flow cytometry and real-time RT-qPCR, respectively. Using xMAP technology, proteomic analysis was also carried out on plasma samples. Unsupervised clustering analysis was used to classify patients based on their immune profiles to further analyze the impact of peripheral immunotypes on patient survival. Significant differences were found between STS patients and HD controls. It was found a contraction of B cells and CD4 T cells compartment, along with decreased expression levels of ICOSLG and CD40LG; a major contribution of suppressor factors, as increased frequency of M-MDSC and memory Tregs, increased expression levels of ARG1, and increased plasma levels of IL-10, soluble VISTA and soluble TIMD-4; and a compromised cytotoxic potential associated with NK and CD8 T cells, namely decreased frequency of CD56dim NK cells, and decreased levels of PRF1, GZMB, and KLRK1. In addition, the patients were classified into three peripheral immunotype groups: "immune-high," "immune-intermediate," and "immune-low." Furthermore, it was found a correlation between these immunotypes and patient survival. Patients classified as "immune-high" exhibited higher levels of immune-related factors linked to cytotoxic/effector activity and longer survival times, whereas patients classified as "immune-low" displayed higher levels of immune factors associated with immunosuppression and shorter survival times. In conclusion, it can be suggested that STS patients have a compromised systemic immunity, and the correlation between immunotypes and survival emphasizes the importance of studying peripheral blood samples in STS. Assessing the peripheral immune response holds promise as a useful method for monitoring and forecasting outcomes in STS.

Characterization of mitochondrial metabolism related molecular subtypes and immune infiltration in colorectal adenocarcinoma

Colorectal adenocarcinoma (COAD) is the most common subtype of colorectal cancer. Due to the imperfect prognosis of COAD, related prognostic factors and possible mechanisms need to be further investigated. During tumor development, mitochondria help tumor cells survive in a variety of ways, so that further screening of mitochondrial metabolism related targets has positive implications for COAD. We screened the mitochondrial metabolism-related genes (MMRG) associated with the COAD prognosis and explored the MMRG-related molecular subtype characteristics of by unsupervised consensus clustering analysis. Using ESTIMATE and ssGSEA algorithms, we evaluated the immunoinfiltration characteristic landscape of different molecular subtypes defined by MMRG. Combining the expression profiles of differentially expressed genes associated with the MMRG subgroup and the survival characteristics of COAD, we constructed an MMRG prognostic model using LASSO-univariate Cox analysis and successfully validated its impact on independently predicting risk stratification of COAD. The potential clinical value of the MMRG score was subsequently evaluated by subgroup immunoinfiltration characteristics and drug susceptibility prediction analysis. We also offer SEC11A as a new potential target for COAD by single-cell sequencing analysis. The effect of SEC11A on the proliferation, invasion abilities and mitochondrial dysfunction of COAD cells was confirmed through in vitro experiments. Our study provides new insights into the role of MMRG and new target for COAD potential intervention.

Evaluation of a Proteomics-Guided Protein Signature for Breast Cancer Detection in Breast Tissue.

The distinction between noncancerous and cancerous breast tissues is challenging in clinical settings, and discovering new proteomics-based biomarkers remains underexplored. Through a pilot proteomic study (discovery cohort), we aimed to identify a protein signature indicative of breast cancer for subsequent validation using six published proteomics/transcriptomics data sets (validation cohorts). Sequential window acquisition of all theoretical (SWATH)-based mass spectrometry revealed 370 differentially abundant proteins between noncancerous tissue and breast cancer. Protein-protein interaction-based networks and enrichment analyses revealed dysregulation in pathways associated with extracellular matrix organization, platelet degranulation, the innate immune system, and RNA metabolism in breast cancer. Through multivariate unsupervised analysis, we identified a four-protein signature (OGN, LUM, DCN, and COL14A1) capable of distinguishing breast cancer. This dysregulation pattern was consistently verified across diverse proteomics and transcriptomics data sets. Dysregulation magnitude was notably higher in poor-prognosis breast cancer subtypes like Basal-Like and HER2 compared to Luminal A. Diagnostic evaluation (receiver operating characteristic (ROC) curves) of the signature in distinguishing breast cancer from noncancerous tissue revealed area under the curve (AUC) ranging from 0.87 to 0.9 with predictive accuracy of 80% to 82%. Upon stratifying, to solely include the Basal-Like/Triple-Negative subtype, the ROC AUC increased to 0.922-0.959 with predictive accuracy of 84.2%-89%. These findings suggest a potential role for the identified signature in distinguishing cancerous from noncancerous breast tissue, offering insights into enhancing diagnostic accuracy.

Blood transcriptome profiling reveals distinct gene networks induced by mRNA vaccination against COVID-19.

Messenger RNA (mRNA) vaccines represent a new class of vaccines that has been shown to be highly effective during the COVID-19 pandemic and that holds great potential for other preventative and therapeutic applications. While it is known that the transcriptional activity of various genes is altered following mRNA vaccination, identifying and studying gene networks could reveal important scientific insights that might inform future vaccine designs. In this study, we conducted an in-depth weighted gene correlation network analysis of the blood transcriptome before and 24h after the second and third vaccination with licensed mRNA vaccines against COVID-19 in humans, following a prime vaccination with either mRNA or ChAdOx1 vaccines. Utilizing this unsupervised gene network analysis approach, we identified distinct modular networks of co-varying genes characterized by either an expressional up- or downregulation in response to vaccination. Downregulated networks were associated with cell metabolic processes and regulation of transcription factors, while upregulated networks were associated with myeloid differentiation, antigen presentation, and antiviral, interferon-driven pathways. Within this interferon-associated network, we identified highly connected hub genes such as STAT2 and RIGI and associated upstream transcription factors, potentially playing important regulatory roles in the vaccine-induced immune response. The expression profile of this network significantly correlated with S1-specific IgG levels at the follow-up visit in vaccinated individuals. Those findings could be corroborated in a second, independent cohort of mRNA vaccine recipients. Collectively, results from this modular gene network analysis enhance the understanding of mRNA vaccines from a systems immunology perspective. Influencing specific gene networks could lead to optimized vaccines that elicit augmented vaccine responses.

Unsupervised news analysis for enhanced high‐frequency food insecurity assessment

AbstractThis article introduces an artificial intelligence (AI)‐based system for forecasting food insecurity in data‐limited settings, employing unsupervised neural networks for topic modeling on news data. Unlike traditional methods, our system operates without relying on expert assumptions about food insecurity factors. Through a case study in Somalia, we show that the method can yield competitive performance, even in the absence of traditional food security indicators such as food prices. This system is valuable in supporting expert assessments of food insecurity, unlocking a wealth of untapped information from news outlets, and offering a path toward more frequent and automated food insecurity monitoring for timely crisis intervention.