Machine learning algorithms reveal unique gene expression profiles in muscle biopsies from patients with different types of myositis

Abstract

ObjectivesMyositis is a heterogeneous family of diseases that includes dermatomyositis (DM), antisynthetase syndrome (AS), immune-mediated necrotising myopathy (IMNM), inclusion body myositis (IBM), polymyositis and overlap myositis. Additional subtypes of myositis can be defined by the presence of myositis-specific autoantibodies (MSAs). The purpose of this study was to define unique gene expression profiles in muscle biopsies from patients with MSA-positive DM, AS and IMNM as well as IBM.MethodsRNA-seq was performed on muscle biopsies from 119 myositis patients with IBM or defined MSAs and 20 controls. Machine learning algorithms were trained on transcriptomic data and recursive feature elimination was used to determine which genes were most useful for classifying muscle biopsies into each type and MSA-defined subtype of myositis.ResultsThe support vector machine learning algorithm classified the muscle biopsies with >90% accuracy. Recursive feature elimination identified genes that are most useful to the machine learning algorithm and that are only overexpressed in one type of myositis. For example, CAMK1G (calcium/calmodulin-dependent protein kinase IG), EGR4 (early growth response protein 4) and CXCL8 (interleukin 8) are highly expressed in AS but not in DM or other types of myositis. Using the same computational approach, we also identified genes that are uniquely overexpressed in different MSA-defined subtypes. These included apolipoprotein A4 (APOA4), which is only expressed in anti-3-hydroxy-3-methylglutaryl-CoA reductase (HMGCR) myopathy, and MADCAM1 (mucosal vascular addressin cell adhesion molecule 1), which is only expressed in anti-Mi2-positive DM.ConclusionsUnique gene expression profiles in muscle biopsies from patients with MSA-defined subtypes of myositis and IBM suggest that different pathological mechanisms underly muscle damage in each of these diseases.

Ultra-efficient sequencing of T Cell receptor repertoires reveals shared responses in muscle from patients with Myositis

“…This enabled the absolute estimation of T cells in each sample, while also providing RNA templates to ensure PCR product generation in samples with few or no infiltrating T cells. Aggregated CD4 and CD8 sequencing reads from RNA-seq analysis [39] of the same samples tightly correlated with a FR3AK-seq based estimate of cellular equivalents (Fig 6a). In subsequent analyses, CDR3 sequences present at levels at or below one cell equivalent per biopsy were considered 'bystander' T cell clones, unlikely to be involved in the disease process.…”

“…IIM patients included those with dermatomyositis (DM, 40), immune-mediated necrotizing myopathy (IMNM, 49), sporadic inclusion body myositis (IBM, 14), and anti-synthetase syndrome (ASyS, 21) ( Table S4 ). All patients provided informed consent and were not selected to be naïve to treatment [39] . All subjects were enrolled in institutional review board (IRB)-approved longitudinal cohorts from the National Institutes of Health (IRB number 91-AR-0196), the Johns Hopkins Myositis Center (IRB number NA_00007454), the Clinic Hospital (Barcelona; IRB number HCB/2015/0479), and the Vall d'Hebron Hospital (Barcelona; IRB number PR (AG) 68/2008).…”

“…IIM patients included those with dermatomyositis (DM, 40), immune-mediated necrotizing myopathy (IMNM, 49), sporadic inclusion body myositis (IBM, 14), and anti-synthetase syndrome (ASyS, 21) (Table S4). All patients provided informed consent and were not selected to be naïve to treatment [39]…”

Lupus or not? SLE Risk Probability Index (SLERPI): a simple, clinician-friendly machine learning-based model to assist the diagnosis of systemic lupus erythematosus

“…Such models can be trained from different kinds of medical or biological data. 13 14 ML has been used for the molecular classification of inflammatory myositis 15 and rheumatoid arthritis, 16 for predicting mortality, 17 response to biological agents 18 and disease activity, 19 whereas less effort has been directed towards diagnosis. 20 21 Building robust computational models that avoid excess complexity represents an important challenge.…”

Correspondence on ‘Machine learning algorithms reveal unique gene expression profiles in muscle biopsies from patients with different types of myositis‘

“…I read the paper by Pinal-Fernandez et al in your journal with great interest. 1 They reported that machine algorithms can be trained on transcriptomic data to classify muscle biopsies from patients with various idiopathic inflammatory myopathies (IIM). I think this classification model can be applied to personalised medicine by targeting the specific molecules.…”