Non-negative tensor factorization workflow for time series biomedical data.

Non-negative tensor factorization (NTF) enables the extraction of a small number of latent components from high-dimensional biomedical data. However, NTF requires many steps, which is a hurdle to implementation. Here, we provide a protocol for TensorLyCV, an easy to run and reproducible NTF analysis pipeline using Snakemake workflow management system and Docker container. Using vaccine adverse reaction data as an example, we describe steps for data processing, tensor decomposition, optimal rank parameter estimation, and visualization of factor matrices. For complete details on the use and execution of this protocol, please refer to Kei Ikeda et al.1.

Detecting time-evolving phenotypic components of adverse reactions against BNT162b2 SARS-CoV-2 vaccine via non-negative tensor factorization.

Symptoms of adverse reactions to vaccines evolve over time, but traditional studies have focused only on the frequency and intensity of symptoms. Here, we attempt to extract the dynamic changes in vaccine adverse reaction symptoms as a small number of interpretable components by using non-negative tensor factorization. We recruited healthcare workers who received two doses of the BNT162b2 mRNA COVID-19 vaccine at Chiba University Hospital and collected information on adverse reactions using a smartphone/web-based platform. We analyzed the adverse-reaction data after each dose obtained for 1,516 participants who received two doses of vaccine. The non-negative tensor factorization revealed four time-evolving components that represent typical temporal patterns of adverse reactions for both doses. These components were differently associated with background factors and post-vaccine antibody titers. These results demonstrate that complex adverse reactions against vaccines can be explained by a limited number of time-evolving components identified by tensor factorization.

A mathematical model for dynamics of soluble form of DNAM-1 as a biomarker for graft-versus-host disease.

DNAM-1 (CD226) is an activating immunoreceptor expressed on T cells and NK cells and involved in the pathogenesis of acute graft-versus-host disease (aGVHD) after allogeneic hematopoietic stem cell transplantation (allo-HSCT). We previously reported that a soluble form of DNAM-1 (sDNAM-1) is generated by shedding from activated T cells. Moreover, higher serum levels of sDNAM-1 in patients before allo-HSCT is a predictive biomarker for the development of aGVHD based on the retrospective univariate and multivariate analyses in allo-HSCT patients. However, it remains unclear how the serum levels of sDNAM-1 are regulated after allo-HSCT and whether they are associated with the development of aGVHD. Here, we constructed a mathematical model to assess the dynamics of sDNAM-1 after allo-HSCT by assuming that there are three types of sDNAM-1 (the first and the second were from alloreactive and non-alloreactive donor lymphocytes, respectively, and the third from recipient lymphocytes). Our mathematical model fitted well to the data set of sDNAM-1 in patients (n = 67) who had undergone allo-HSCT and suggest that the high proportion of the first type of sDNAM-1 to the total of the first and second types is associated with high risk of the development of severe aGVHD. Thus, sDNAM-1 after allo-HSCT can be a biomarker for the development of aGVHD.