Motivation Co-expression networks are a powerful gene expression analysis method to study how genes co-express together in clusters with functional coherence that usually resemble specific cell type behaviour for the genes involved. They can be applied to bulk-tissue gene expression profiling and assign function, and usually cell type specificity, to a high percentage of the gene pool used to construct the network. One of the limitations of this method is that each gene is predicted to play a role in a specific set of coherent functions in a single cell type (i.e. at most we get a single <gene, function, cell type> for each gene). We present here GMSCA (Gene Multifunctionality Secondary Co-expression Analysis), a software tool that exploits the co-expression paradigm to increase the number of functions and cell types ascribed to a gene in bulk-tissue co-expression networks. Results We applied GMSCA to 27 co-expression networks derived from bulk-tissue gene expression profiling of a variety of brain tissues. Neurons and glial cells (microglia, astrocytes and oligodendrocytes) were considered the main cell types. Applying this approach, we increase the overall number of predicted triplets <gene, function, cell type> by 46.73%. Moreover, GMSCA predicts that the SNCA gene, traditionally associated to work mainly in neurons, also plays a relevant function in oligodendrocytes. Availability The tool is available at GitHub, https://github.com/drlaguna/GMSCA as open-source software. Supplementary information Supplementary data are available at Bioinformatics online.

Mild Cognitive Impairment (MCI) is a phase that can precede Alzheimer's Disease (AD). To better understand the molecular mechanisms underlying conversion from MCI to AD, we proposed a multiomics machine learning pipeline (four algorithms) to identify key pathways. Data consisted of metabolites (n=540) and proteins (n=3630) measured in blood plasma coupled with standard clinical tests (n=26). The cohort comprised 230 controls, 386 MCI participants and 184 AD-type dementia participants. Multiclass models showed that oleamide, MMSE and the priority language Z-score were the most relevant variables. Oleamide was increased in the MCI group and further increased in converters (both P<0.0001). In-vitro disease-associated microglia were able to synthesize oleamide and excrete it in vesicles. MCI conversion models showed pTau, tTau and JPH3, CFP, synuclein and PI15 proteins as the most relevant. This study uncovered molecular pathways in MCI conversion involved in inflammation (oleamide, CFP), neuronal regulation (JPH3, SNCA) and protein degradation (PI15).

Gene set based phenotype enrichment analysis (detecting phenotypic terms that emerge as significant in a set of genes) can improve the rate of genetic diagnoses amongst other research purposes. To facilitate diverse phenotype analysis, we developed PhenoExam, a freely available R package for tool developers and a web interface for users, which performs: (1) phenotype and disease enrichment analysis on a gene set; (2) measures statistically significant phenotype similarities between gene sets and (3) detects significant differential phenotypes or disease terms across different databases. PhenoExam achieves these tasks by integrating databases or resources such as the HPO, MGD, CRISPRbrain, CTD, ClinGen, CGI, OrphaNET, UniProt, PsyGeNET, and Genomics England Panel App. PhenoExam accepts both human and mouse genes as input. We developed PhenoExam to assist a variety of users, including clinicians, computational biologists and geneticists. It can be used to support the validation of new gene-to-disease discoveries, and in the detection of differential phenotypes between two gene sets (a phenotype linked to one of the gene set but no to the other) that are useful for differential diagnosis and to improve genetic panels. We validated PhenoExam performance through simulations and its application to real cases. We demonstrate that PhenoExam is effective in distinguishing gene sets or Mendelian diseases with very similar phenotypes through projecting the disease-causing genes into their annotation-based phenotypic spaces. We also tested the tool with early onset Parkinson's disease and dystonia genes, to show phenotype-level similarities but also potentially interesting differences. More specifically, we used PhenoExam to validate computationally predicted new genes potentially associated with epilepsy. Therefore, PhenoExam effectively discovers links between phenotypic terms across annotation databases through effective integration. The R package is available at https://github.com/alexcis95/PhenoExam and the Web tool is accessible at https://snca.atica.um.es/PhenoExamWeb/.

Background Gene set enrichment analysis (detecting phenotypic terms that emerge as significant in a set of genes) plays an important role in bioinformatics focused on diseases of genetic basis. To facilitate phenotype-oriented gene set analysis, we developed PhenoExam, a freely available R package for tool developers and a web interface for users, which performs: (1) phenotype and disease enrichment analysis on a gene set; (2) measures statistically significant phenotype similarities between gene sets and (3) detects significant differential phenotypes or disease terms across different databases. Results PhenoExam generates sensitive and accurate phenotype enrichment analyses. It is also effective in segregating gene sets or Mendelian diseases with very similar phenotypes. We tested the tool with two similar diseases (Parkinson and dystonia), to show phenotype-level similarities but also potentially interesting differences. Moreover, we used PhenoExam to validate computationally predicted new genes potentially associated with epilepsy. Conclusions We developed PhenoExam, a freely available R package and Web application, which performs phenotype enrichment and disease enrichment analysis on gene set G, measures statistically significant phenotype similarities between pairs of gene sets G and G′ and detects statistically significant exclusive phenotypes or disease terms, across different databases. We proved with simulations and real cases that it is useful to distinguish between gene sets or diseases with very similar phenotypes. Github R package URL is https://github.com/alexcis95/PhenoExam. Shiny App URL is https://alejandrocisterna.shinyapps.io/phenoexamweb/.

Motivation: gene co-expression networks have been widely applied to identify critical genes and pathways for neurodegenerative diseases such as Parkinson's and Alzheimer's disease. Now, with the advent of single-cell RNA-sequencing, we have the opportunity to create cell-type specific gene co-expression networks. However, single-cell RNA-sequencing data is characterized by its sparsity, amongst some other issues raised by this new type of data. Results: We present scCoExpNets, a framework for the discovery and analysis of cell-type specific gene coexpression networks (GCNs) from single-cell RNA-seq data. We propose a new strategy to address the problem of sparsity, named iterative pseudo-cell identification. It consists of adding the gene expression of pairs of cells that belong to the same individual and the same cell-type while the number of cells is over 200, thus creating multiple matrices and multiple scGCNs for the same cell-type, all of them seen as alternative and complementary views of the same phenomena. We applied this new tool on a snRNA-seq dataset human post-mortem substantia nigra pars compacta tissue of 13 controls and 14 Parkinson's disease (PD) cases (18 males and 9 females) with 30-99 years. We show that one of the hypotheses that support the selective vulnerability of dopaminergic neurons in PD, the iron accumulation, is sustained in our dopaminergic neurons network models. Moreover, after successive pseudo-celluling iterations, the gene groups sustaining this hypothesis remain intact. At the same time, this pseudo-celulling strategy also allows us to discover genes whose grouping changes considerably throughout the iterations and provides new insights. Finally, since some of our models were correlated with diagnosis and age at the same time, we also developed our own framework to create covariate-specific GCNs, called CovCoExpNets. We applied this new software to our snRNA-seq dataset and we identified 11 age-specific genes and 5 diagnosis-specific genes which do not overlap. Availability and implementation: The CoExpNets implementations are available as R packages, scCoExpNets package, for creating single-cell GCNs and CovCoExpNets, for creating covariate-specific GCNs. Users can either download the development version via github https://github.com/aliciagp/scCoExpNets and https://github.com/aliciagp/CovCoExpNets Contact: alicia.gomez1@um.es Supplementary information: supplementary data is available online. Keywords: weighted gene co-expression networks, single-nucleus RNA-sequencing, sparsity, pseudo-cells, Parkinson's disease, selective vulnerability, dopaminergic neurons, lasso regression