Amplifying pathological detection in EEG signaling pathways through cross-dataset transfer learning.

Pathology diagnosis based on EEG signals and decoding brain activity holds immense importance in understanding neurological disorders. With the advancement of artificial intelligence methods and machine learning techniques, the potential for accurate data-driven diagnoses and effective treatments has grown significantly. However, applying machine learning algorithms to real-world datasets presents diverse challenges at multiple levels. The scarcity of labeled data, especially in low regime scenarios with limited availability of real patient cohorts due to high costs of recruitment, underscores the vital deployment of scaling and transfer learning techniques. In this study, we explore a real-world pathology classification task to highlight the effectiveness of data and model scaling and cross-dataset knowledge transfer. As such, we observe varying performance improvements through data scaling, indicating the need for careful evaluation and labeling. Additionally, we identify the challenges of possible negative transfer and emphasize the significance of some key components to overcome distribution shifts and potential spurious correlations and achieve positive transfer. We see improvement in the performance of the target model on the target (NMT) datasets by using the knowledge from the source dataset (TUAB) when a low amount of labeled data was available. Our findings demonstrated that a small and generic model (e.g. ShallowNet) performs well on a single dataset, however, a larger model (e.g. TCN) performs better in transfer learning when leveraging a larger and more diverse dataset.

A mathematical model of protein subunits COVID-19 vaccines.

We consider a general mathematical model for protein subunit vaccine with a focus on the MF59-adjuvanted spike glycoprotein-clamp vaccine for SARS-CoV-2, and use the model to study immunological outcomes in the humoral and cell-mediated arms of the immune response from vaccination. The mathematical model is fit to vaccine clinical trial data. We elucidate the role of Interferon-γ and Interleukin-4 in stimulating the immune response of the host. Model results, and results from a sensitivity analysis, show that a balance between the TH1 and TH2 arms of the immune response is struck, with the TH1 response being dominant. The model predicts that two-doses of the vaccine at 28 days apart will result in approximately 85% humoral immunity loss relative to peak immunity approximately 6 months post dose 1.

Long-term durability of immune responses to the BNT162b2 and mRNA-1273 vaccines based on dosage, age and sex.

The lipid nanoparticle (LNP)-formulated mRNA vaccines BNT162b2 and mRNA-1273 are a widely adopted multi vaccination public health strategy to manage the COVID-19 pandemic. Clinical trial data has described the immunogenicity of the vaccine, albeit within a limited study time frame. Here, we use a within-host mathematical model for LNP-formulated mRNA vaccines, informed by available clinical trial data from 2020 to September 2021, to project a longer term understanding of immunity as a function of vaccine type, dosage amount, age, and sex. We estimate that two standard doses of either mRNA-1273 or BNT162b2, with dosage times separated by the company-mandated intervals, results in individuals losing more than 99% humoral immunity relative to peak immunity by 8 months following the second dose. We predict that within an 8 month period following dose two (corresponding to the original CDC time-frame for administration of a third dose), there exists a period of time longer than 1 month where an individual has lost more than 99% humoral immunity relative to peak immunity, regardless of which vaccine was administered. We further find that age has a strong influence in maintaining humoral immunity; by 8 months following dose two we predict that individuals aged 18-55 have a four-fold humoral advantage compared to aged 56-70 and 70+ individuals. We find that sex has little effect on the immune response and long-term IgG counts. Finally, we find that humoral immunity generated from two low doses of mRNA-1273 decays at a substantially slower rate relative to peak immunity gained compared to two standard doses of either mRNA-1273 or BNT162b2. Our predictions highlight the importance of the recommended third booster dose in order to maintain elevated levels of antibodies.

Longitudinal Assessment of SARS-CoV-2-Specific T Cell Cytokine-Producing Responses for 1 Year Reveals Persistence of Multicytokine Proliferative Responses, with Greater Immunity Associated with Disease Severity.

Cell-mediated immunity is critical for long-term protection against most viral infections, including coronaviruses. We studied 23 severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-infected survivors over a 1-year post-symptom onset (PSO) interval by ex vivo cytokine enzyme-linked immunosorbent spot assay (ELISpot) assay. All subjects demonstrated SARS-CoV-2-specific gamma interferon (IFN-γ), interleukin 2 (IL-2), and granzyme B (GzmB) T cell responses at presentation, with greater frequencies in severe disease. Cytokines, mainly produced by CD4+ T cells, targeted all structural proteins (nucleocapsid, membrane, and spike) except envelope, with GzmB and IL-2 greater than IFN-γ. Mathematical modeling predicted that (i) cytokine responses peaked at 6 days for IFN-γ, 36 days for IL-2, and 7 days for GzmB, (ii) severe illness was associated with reduced IFN-γ and GzmB but increased IL-2 production rates, and (iii) males displayed greater production of IFN-γ, whereas females produced more GzmB. Ex vivo responses declined over time, with persistence of IL-2 in 86% and of IFN-γ and GzmB in 70% of subjects at a median of 336 days PSO. The average half-life of SARS-CoV-2-specific cytokine-producing cells was modeled to be 139 days (~4.6 months). Potent T cell proliferative responses persisted throughout observation, were CD4 dominant, and were capable of producing all 3 cytokines. Several immunodominant CD4 and CD8 epitopes identified in this study were shared by seasonal coronaviruses or SARS-CoV-1 in the nucleocapsid and membrane regions. Both SARS-CoV-2-specific CD4+ and CD8+ T cell clones were able to kill target cells, though CD8 tended to be more potent. IMPORTANCE Our findings highlight the relative importance of SARS-CoV-2-specific GzmB-producing T cell responses in SARS-CoV-2 control and shared CD4 and CD8 immunodominant epitopes in seasonal coronaviruses or SARS-CoV-1, and they indicate robust persistence of T cell memory at least 1 year after infection. Our findings should inform future strategies to induce T cell vaccines against SARS-CoV-2 and other coronaviruses.

A machine learning approach to differentiate between COVID-19 and influenza infection using synthetic infection and immune response data.

Data analysis is widely used to generate new insights into human disease mechanisms and provide better treatment methods. In this work, we used the mechanistic models of viral infection to generate synthetic data of influenza and COVID-19 patients. We then developed and validated a supervised machine learning model that can distinguish between the two infections. Influenza and COVID-19 are contagious respiratory illnesses that are caused by different pathogenic viruses but appeared with similar initial presentations. While having the same primary signs COVID-19 can produce more severe symptoms, illnesses, and higher mortality. The predictive model performance was externally evaluated by the ROC AUC metric (area under the receiver operating characteristic curve) on 100 virtual patients from each cohort and was able to achieve at least AUC = 91% using our multiclass classifier. The current investigation highlighted the ability of machine learning models to accurately identify two different diseases based on major components of viral infection and immune response. The model predicted a dominant role for viral load and productively infected cells through the feature selection process.

Analysis of Host Immunological Response of Adenovirus-Based COVID-19 Vaccines.

During the SARS-CoV-2 global pandemic, several vaccines, including mRNA and adenovirus vector approaches, have received emergency or full approval. However, supply chain logistics have hampered global vaccine delivery, which is impacting mass vaccination strategies. Recent studies have identified different strategies for vaccine dose administration so that supply constraints issues are diminished. These include increasing the time between consecutive doses in a two-dose vaccine regimen and reducing the dosage of the second dose. We consider both of these strategies in a mathematical modeling study of a non-replicating viral vector adenovirus vaccine in this work. We investigate the impact of different prime-boost strategies by quantifying their effects on immunological outcomes based on simple system of ordinary differential equations. The boost dose is administered either at a standard dose (SD) of 1000 or at a low dose (LD) of 500 or 250 vaccine particles. Results show dose-dependent immune response activity. Our model predictions show that by stretching the prime-boost interval to 18 or 20, in an SD/SD or SD/LD regimen, the minimum promoted antibody (Nab) response will be comparable with the neutralizing antibody level measured in COVID-19 recovered patients. Results also show that the minimum stimulated antibody in SD/SD regimen is identical with the high level observed in clinical trial data. We conclude that an SD/LD regimen may provide protective capacity, which will allow for conservation of vaccine doses.

Multiomics Characterization of Preterm Birth in Low- and Middle-Income Countries.

Importance: Worldwide, preterm birth (PTB) is the single largest cause of deaths in the perinatal and neonatal period and is associated with increased morbidity in young children. The cause of PTB is multifactorial, and the development of generalizable biological models may enable early detection and guide therapeutic studies. Objective: To investigate the ability of transcriptomics and proteomics profiling of plasma and metabolomics analysis of urine to identify early biological measurements associated with PTB. Design, Setting, and Participants: This diagnostic/prognostic study analyzed plasma and urine samples collected from May 2014 to June 2017 from pregnant women in 5 biorepository cohorts in low- and middle-income countries (LMICs; ie, Matlab, Bangladesh; Lusaka, Zambia; Sylhet, Bangladesh; Karachi, Pakistan; and Pemba, Tanzania). These cohorts were established to study maternal and fetal outcomes and were supported by the Alliance for Maternal and Newborn Health Improvement and the Global Alliance to Prevent Prematurity and Stillbirth biorepositories. Data were analyzed from December 2018 to July 2019. Exposures: Blood and urine specimens that were collected early during pregnancy (median sampling time of 13.6 weeks of gestation, according to ultrasonography) were processed, stored, and shipped to the laboratories under uniform protocols. Plasma samples were assayed for targeted measurement of proteins and untargeted cell-free ribonucleic acid profiling; urine samples were assayed for metabolites. Main Outcomes and Measures: The PTB phenotype was defined as the delivery of a live infant before completing 37 weeks of gestation. Results: Of the 81 pregnant women included in this study, 39 had PTBs (48.1%) and 42 had term pregnancies (51.9%) (mean [SD] age of 24.8 [5.3] years). Univariate analysis demonstrated functional biological differences across the 5 cohorts. A cohort-adjusted machine learning algorithm was applied to each biological data set, and then a higher-level machine learning modeling combined the results into a final integrative model. The integrated model was more accurate, with an area under the receiver operating characteristic curve (AUROC) of 0.83 (95% CI, 0.72-0.91) compared with the models derived for each independent biological modality (transcriptomics AUROC, 0.73 [95% CI, 0.61-0.83]; metabolomics AUROC, 0.59 [95% CI, 0.47-0.72]; and proteomics AUROC, 0.75 [95% CI, 0.64-0.85]). Primary features associated with PTB included an inflammatory module as well as a metabolomic module measured in urine associated with the glutamine and glutamate metabolism and valine, leucine, and isoleucine biosynthesis pathways. Conclusions and Relevance: This study found that, in LMICs and high PTB settings, major biological adaptations during term pregnancy follow a generalizable model and the predictive accuracy for PTB was augmented by combining various omics data sets, suggesting that PTB is a condition that manifests within multiple biological systems. These data sets, with machine learning partnerships, may be a key step in developing valuable predictive tests and intervention candidates for preventing PTB.

Multiomics modeling of the immunome, transcriptome, microbiome, proteome and metabolome adaptations during human pregnancy.

Motivation: Multiple biological clocks govern a healthy pregnancy. These biological mechanisms produce immunologic, metabolomic, proteomic, genomic and microbiomic adaptations during the course of pregnancy. Modeling the chronology of these adaptations during full-term pregnancy provides the frameworks for future studies examining deviations implicated in pregnancy-related pathologies including preterm birth and preeclampsia. Results: We performed a multiomics analysis of 51 samples from 17 pregnant women, delivering at term. The datasets included measurements from the immunome, transcriptome, microbiome, proteome and metabolome of samples obtained simultaneously from the same patients. Multivariate predictive modeling using the Elastic Net (EN) algorithm was used to measure the ability of each dataset to predict gestational age. Using stacked generalization, these datasets were combined into a single model. This model not only significantly increased predictive power by combining all datasets, but also revealed novel interactions between different biological modalities. Future work includes expansion of the cohort to preterm-enriched populations and in vivo analysis of immune-modulating interventions based on the mechanisms identified. Availability and implementation: Datasets and scripts for reproduction of results are available through: https://nalab.stanford.edu/multiomics-pregnancy/. Supplementary information: Supplementary data are available at Bioinformatics online.