Type 1 Gaucher disease (GD1) is a rare, autosomal recessive disorder caused by glucocerebrosidase deficiency. Skeletal manifestations represent one of the most debilitating and potentially irreversible complications of GD1. Although imaging studies are the gold standard, early diagnostic/prognostic tools, such as molecular biomarkers, are needed for the rapid management of skeletal complications. This study aimed to identify potential protein biomarkers capable of predicting the early diagnosis of bone skeletal complications in GD1 patients using artificial intelligence. An in silico study was performed using the novel Therapeutic Performance Mapping System methodology to construct mathematical models of GD1-associated complications at the protein level. Pathophysiological characterization was performed before modeling, and a data science strategy was applied to the predicted protein activity for each protein in the models to identify classifiers. Statistical criteria were used to prioritize the most promising candidates, and 18 candidates were identified. Among them, PDGFB, IL1R2, PTH and CCL3 (MIP-1α) were highlighted due to their ease of measurement in blood. This study proposes a validated novel tool to discover new protein biomarkers to support clinician decision-making in an area where medical needs have not yet been met. However, confirming the results using in vitro and/or in vivo studies is necessary.

Metachromatic leukodystrophy (MLD) is a rare, autosomal recessive lysosomal storage disease. Deficient activity of arylsulfatase A causes sulfatides to accumulate in cells of different tissues, including those in the central and peripheral nervous systems, leading to progressive demyelination and neurodegeneration. Although there is some association between specific arylsulfatase A alleles and disease severity, genotype-phenotype correlations are not fully understood. We aimed to identify biomarker candidates of early tissue damage in MLD using a modeling approach based on systems biology. A review of the literature was performed in an initial disease characterization step, allowing identification of pathophysiological processes involved in MLD and proteins relating to these processes. Three mathematical models were generated to simulate different stages of MLD at the molecular level: an early pro-inflammatory stage model (including only processes considered to be active in the early stages of disease), a pre-demyelination stage model (including additional processes that are active after some disease progression), and a demyelination stage model (in which all pathophysiological processes are active). The models evaluated 3457 proteins of interest, individually and by pairs through data mining techniques, applying five filters to prioritize biomarkers that could differentiate between the models. Sixteen potential biomarkers were identified, including effectors relating to mitochondrial dysfunction, remyelination, and neurodegeneration. The findings were corroborated in a gene expression data set from T lymphocytes of patients with MLD; all candidates formed combinations that were able to distinguish patients with MLD from controls, and all but one candidate distinguished late-infantile MLD from juvenile MLD as part of a combinatorial biomarker pair. In particular, pro-neuregulin-1 appeared as differential on all comparisons (patients with MLD vs controls and within clinical subtypes); casein kinase II subunit alpha was detected as a potential individual marker within clinical subtypes. These findings provide a panel of biomarker candidates suitable for experimental validation and highlight the utility of mathematical models to identify biomarker candidates of early tissue damage in MLD with a high degree of accuracy and sensitivity.