Nearest neighbor (NN) imputation algorithms are efficient methods to fill in missing data where each missing value on some records is replaced by a value obtained from related cases in the whole set of records. Besides the capability to substitute the missing data with plausible values that are as close as possible to the true value, imputation algorithms should preserve the original data structure and avoid to distort the distribution of the imputed variable. Despite the efficiency of NN algorithms little is known about the effect of these methods on data structure. Simulation on synthetic datasets with different patterns and degrees of missingness were conducted to evaluate the performance of NN with one single neighbor (1NN) and with k neighbors without (kNN) or with weighting (wkNN) in the context of different learning frameworks: plain set, reduced set after ReliefF filtering, bagging, random choice of attributes, bagging combined with random choice of attributes (Random-Forest-like method). Whatever the framework, kNN usually outperformed 1NN in terms of precision of imputation and reduced errors in inferential statistics, 1NN was however the only method capable of preserving the data structure and data were distorted even when small values of k neighbors were considered; distortion was more severe for resampling schemas. The use of three neighbors in conjunction with ReliefF seems to provide the best trade-off between imputation error and preservation of the data structure. The very same conclusions can be drawn when imputation experiments were conducted on the single proton emission computed tomography (SPECTF) heart dataset after introduction of missing data completely at random.

Intestinal microbiota has been associated with systemic autoimmune diseases, yet the functional consequences of these associations are elusive. We characterized the fecal microbiota (16S rRNA gene amplification and sequencing) and the plasma metabolome (high‐performance liquid chromatography coupled to mass spectrometry) in 59 patients with systemic sclerosis (SSc) and 28 healthy controls (HCs). Microbial and metabolic data were cross‐correlated to find meaningful associations after extensive data mining analysis and internal validation. Our data show that a reduced model of nine bacteria is capable of differentiating HCs from SSc patients. SSc gut microbiota is characterized by a reduction in protective butyrate‐producing bacteria and by an increase in proinflammatory noxious genera, especially Desulfovibrio. From the metabolic point of view, a multivariate model with 17 metabolite intermediates well distinguished cases from controls. The most interesting peaks we found were identified as glycerophospholipid metabolites and benzene derivatives. The microbial and metabolic data showed significant interactions between Desulfovibrio and alpha‐N‐phenylacetyl‐l‐glutamine and 2,4‐dinitrobenzenesulfonic acid. Our data suggest that in SSc, intestinal microbiota is characterized by proinflammatory alterations subtly entwined with the metabolic state. Desulfovibrio is a relevant actor in gut dysbiosis that may promote intestinal damage and influence amino acid metabolism.

Objectives. Interstitial lung disease (ILD) profoundly affects the health-related quality of life (HRQoL) in patients with systemic sclerosis (SSc). We tested the validity of the Saint George's Respiratory Questionnaire (SGRQ), a lung-specific HRQoL-evaluation tool, in a population of SSc patients with ILD.

Objective. To improve knowledge of vasculopathy in SSc through the assessment of serum levels of circulating angiogenetic and endothelial dysfunction markers in patients at different stages of the disease.

The term ‘undifferentiated connective tissue disease’ (UCTD) is generally used to describe clinical entities characterised by clinical and serological manifestations of systemic autoimmune diseases but not fulfilling the criteria for defined connective tissue diseases (CTDs). In this narrative review, we summarise the results of a systematic literature research, which was performed as part of the ERN ReCONNET project, aimed at evaluating existing clinical practice guidelines (CPGs) or recommendations. No specific CPG on UCTD were found, potential areas of intervention are absence of a consensus definition of UCTD, need for specific monitoring and therapeutic protocols, stratification of UCTD based on the risk of developing a defined CTD and preventive measure for the future development of a more severe condition. Patients feel uncertainty regarding the name of the disease and feel the need of a better education and understanding of these conditions and its possible changes over time.