Biology subjects

Botia, J. A.

Publications and source records attributed to Botia, J. A..

At least 19 recordsLinked to original sources

bioRxiv2026

The Almond Expression Atlas: A comprehensive multi-tissue transcriptomic resource for almond gene expression

Background: Improving crop adaptation and enhancing agronomic and quality traits requires a deeper understanding of gene expression regulation, for which expression atlases are valuable tools. Almond (Prunus dulcis (Mill.) D.A. Webb) is one of the most economically important tree nut crops worldwide. However, an integrated transcriptomic resource for this species is still lacking. Available data remain fragmented across independent studies, limiting their comparability and reuse. Results: Here, we present the first almond gene expression atlas: a comprehensive multi-tissue transcriptomic resource integrating publicly available RNA sequencing data. A total of 142 high-quality untreated and control samples from 17 independent projects were curated through a metadata standardization process and analyzed using a unified bioinformatics pipeline. Standardized expression profiles were generated for 25,225 annotated genes in the Prunus dulcis Texas v2.0 reference genome. The atlas integrates two complementary expression approaches: a median TPM approach capturing quantitative expression levels and a rank-based strategy implemented using the Bgee expression calling method, designed to improve comparability across heterogeneous datasets. This design enables the identification of tissue-specific candidate genes and the exploration of expression patterns across tissues and developmental stages. Its applicability is demonstrated through two use cases, highlighting its ability to identify biologically meaningful gene sets associated with bud dormancy transitions and to prioritize candidate reference genes for RT-qPCR normalization in almond. Conclusions: The Almond Expression Atlas v1.0 is publicly available through FruitDBase (fruitdbase.csic.es), providing an interactive platform for gene expression exploration and visualization. This resource supports functional genomics studies and facilitates candidate gene discovery in almond, while ensuring data traceability and promoting data reusability.

bioinformatics↗

bioRxiv2026

Antimicrobial peptide databases and prediction tools: Toward a standard evaluation framework

Antimicrobial resistance (AMR) has a profound impact on animal and human health and is associated with substantial morbidity, mortality and public health costs. There is a clear need to develop novel, effective antibiotic agents, which can overcome the current AMR crisis. Antimicrobial peptides (AMPs) may offer such a solution and have attracted growing attention for their potential to combat AMR. In parallel, the growing availability of peptide sequences in public databases has stimulated the development of numerous machine learning and deep learning tools to predict antimicrobial activity computationally. However, it remains unclear how reliably these tools can be compared, as existing studies often rely on heterogeneous datasets and inconsistent evaluation protocols that may lead to data leakage and inflated performance estimates. This raises a central question: what evaluation criteria and benchmark resources are needed to enable fair, reproducible, and biologically meaningful assessment of AMP prediction tools? We address this question by focusing specifically on antibacterial peptides (ABPs). We first provide an overview of AMP databases relevant to antibacterial activity and compare their content, redundancy, and experimental metadata. We then critically assess existing computational tools for ABP prediction, highlighting key limitations related to dataset construction, affinity to certain sequences, data leakage, and inconsistent performance reporting. Based on these limitations, we propose a reference evaluation framework designed to improve comparability, reproducibility, and practical utility in ABP prediction. Finally, we provide targeted recommendations for AMP databases and future tool development to support more robust progress in the computational discovery of ABPs.

bioinformatics↗

bioRxiv2025

go3: A Fast and Lightweight Library for Semantic Similarity of GO Terms and Genes

MotivationCalculation of semantic similarity of Gene Ontology (GO) term subsets is a fundamental task in functional genomics, comparative studies, and biomedical data integration. Existing tools, primarily in Python or R, often face severe limitations in performance when scaling to large annotation datasets. ResultsWe present go3, the first high-performance, Python-compatible library written in Rust that supports multiple semantic similarity metrics for GO terms and genes. go3 supports both pairwise and batch computations, optimized using Rusts parallelism and memory safety. Compared to GOATOOLS, the state of the art, it achieves up to 5x speedup and 25x lower memory footprint when loading the GO ontology and gene annotations, and up to x 10 3 speedup when calculating semantic similarities between genes, while preserving output compatibility. Availability and implementationgo3 is implemented in Rust and available through Python 3. It is accessible in GitHub: (https://github.com/Mellandd/GO3). Contactjoseluis.mellinaa@um.es

bioinformatics↗

bioRxiv2025

Transcriptomic analysis of repeat expansion-ataxias uncovers distinct non-neuronal cell type-specific signatures of disease across the human brain

Hereditary ataxias are a heterogeneous group of neurogenetic conditions characterised by the clinical syndrome of progressive loss of coordination from neurodegeneration of the cerebellum. A commonality across the most prevalent ataxias is the underlying disease mechanism secondary to expansions of short tandem DNA repeats. There is currently an incomplete understanding of the pathogenic mechanisms of these repeat expansion disorders, a core feature of which revolves around RNA-dysregulation. In this study, we used both bulk and single nuclear RNA-sequencing to study post-mortem brain tissue of human donors with a range of repeat-expansion ataxias to reveal further mechanistic insights. We compared post-mortem paired cerebellar and frontal cortex tissue bulk RNA-sequencing data from 23 ataxia patients and 22 sex-, age-matched controls from two brain banks (spinocerebellar ataxia (SCA)1, SCA2, SCA6, SCA7, SCA17, Friedreichs ataxia (FRDA), and 7 cases with unknown molecular diagnoses). We analysed bulk RNA-sequencing data for transcript usage, differential and cell-type-specific expression to transcriptomically profile these diseases. We also generated single nuclear RNA-sequencing data of the cerebellum from donors with SCA1, SCA2, SCA6 and FRDA to decipher changes in cell type proportions in the disease state. Using this approach, we found that: (i) despite the commonalities in the genetics of ataxia, there were components of their transcriptional signatures which were distinct; (ii) there were extensive transcriptional changes evident not only in the cerebellum but also the frontal cortex in ataxia cases; (iii) activation of immune and inflammatory pathways, as well as involvement of non-neuronal cell types was a feature of all ataxias to a lesser or greater extent. This study provides a novel resource to understand the mechanisms of disease in ataxia. Furthermore, taken together, these results highlight immune pathways and the role of non-neuronal cell types as early and potentially important therapeutic targets. These findings provide a map of transcriptomic changes in ataxia to further understanding of the underlying pathogenesis.

neuroscience↗

bioRxiv2024

Long-read transcriptomic identification of synaptic adaptation to amyloid pathology in the App(NL-G-F) knock-in mouse model of the earliest phase of Alzheimer's disease

Genome-wide association studies (GWAS) have identified a transcriptional network of Alzheimers disease (AD) risk genes that are primarily expressed in microglia and are associated with AD pathology. However, traditional short-read sequencers have limited our ability to fully characterize how GWAS variants exert their effects on gene expression regulation or alternative splicing in response to the pathology, particularly resulting in inaccurate detection of splicing. To address this gap, we utilized long-read RNA-sequencing (RNA-seq) in the AppNL-G-F knock-in mouse model to identify changes in splicing and novel transcript isoforms in response to amyloid-{beta}. We show that long-read RNA-seq can recapitulate the expected induction of microglial expressed risk genes such as Trem2 in response to amyloid-{beta} at 9 months of age associated with ageing-dependent deficiencies in spatial short-term memory in the AppNL-G-F knock-in mice. Our results not only identified novel splicing events and transcript isoform abundance in genes associated with AD, but also revealed the complex regulation of gene expression through splicing in response to amyloid plaques. Surprisingly, the regulation of alternative splicing in response to amyloid was seen in genes previously not identified as AD risk genes, expressed in microglia, neurons and oligodendrocytes, and included genes such as Syngr1 that modulate synaptic physiology. We saw alternative splicing in genes such as Ctsa, Clta, Dennd2a, Irf9 and Smad4 in mice in response to amyloid, and the orthologues of these genes also showed transcript usage changes in human AD brains. Our data suggests a model whereby induction of AD risk gene expression associated with microglial proliferation and activation is concomitant with alternative splicing in a different class of genes expressed by microglia and neurons, which act to adapt or preserve synaptic activity in response to amyloid-{beta} during early stages of the disease. Our study provides new insights into the mechanisms and effects of the regulation of genes associated with amyloid pathology, which may ultimately enable better disease diagnosis, and improved tracking of disease progression. Additionally, our findings identify new therapeutic avenues for treatment of AD.

genetics↗

bioRxiv2023

Splicing accuracy varies across human introns, tissues and age

Alternative splicing impacts most multi-exonic human genes. Inaccuracies during this process may have an important role in ageing and disease. Here, we investigated mis-splicing using RNA-sequencing data from ~14K control samples and 42 human body sites, focusing on split reads partially mapping to known transcripts in annotation. We show that mis-splicing occurs at different rates across introns and tissues and that these splicing inaccuracies are primarily affected by the abundance of core components of the spliceosome assembly and its regulators. Using publicly available data on short-hairpin RNA-knockdowns of numerous spliceosomal components and related regulators, we found support for the importance of RNA-binding proteins in mis-splicing. We also demonstrated that age is positively correlated with mis-splicing, and it affects genes implicated in neurodegenerative diseases. This in-depth characterisation of mis-splicing can have important implications for our understanding of the role of splicing inaccuracies in human disease and the interpretation of long-read RNA-sequencing data.

genetics↗

bioRxiv2023

GenoDrawing: An autoencoder framework for image prediction from SNP markers

Advancements in genome sequencing have facilitated whole genome characterization of numerous plant species, providing an abundance of genotypic data for genomic analysis. Genomic selection and neural networks, particularly deep learning, have been developed to predict complex traits from dense genotypic data. Autoencoders, a neural network model to extract features from images in an unsupervised manner, has proven to be useful for plant phenotyping. This study introduces an autoencoder framework, GenoDrawing, for predicting and retrieving apple images from a low-depth single nucleotide polymorphism (SNP) array, potentially useful in predicting traits that are difficult to define. GenoDrawing demonstrated proficiency in its task while using a small dataset of shape-related SNPs, and multiple experiments were conducted to evaluate the impact of SNP selection and shape relation. Results indicated that the correct relationship of SNPs with visual traits had a significant impact on the generated images, consistent with biological interpretation. While using significant SNPs is crucial, incorporating additional, unrelated SNPs results in performance degradation for simple NN architectures that cannot easily identify the most important inputs. The proposed GenoDrawing method is a practical framework for exploring genomic prediction in fruit tree phenotyping, particularly beneficial for small to medium breeding companies to predict economically significant heritable traits. Although GenoDrawing has limitations, it sets the groundwork for future research in image prediction from genomic markers. Future studies should focus on using stronger models for image reproduction, SNP information extraction, and improved dataset balance in terms of shape for more precise outcomes.

plant biology↗

bioRxiv2023

The chromatin modulating NSL complex regulates genes and pathways genetically linked to Parkinsons disease

Genetic variants conferring risk for Parkinsons disease have been highlighted through genome-wide association studies, yet exploration of their specific disease mechanisms is lacking. Two Parkinsons disease candidate genes, KAT8 and KANSL1, identified through genome-wide studies and a PINK1-mitophagy screen, encode part of the histone acetylating non-specific lethal complex. This complex localises to the nucleus, where it has a role in transcriptional activation, and to mitochondria, where it has been suggested to have a role in mitochondrial transcription. In this study, we sought to identify whether the non-specific lethal complex has potential regulatory relationships with other genes associated with Parkinsons disease in human brain. Correlation in the expression of non-specific lethal genes and Parkinsons disease-associated genes was investigated in primary gene co-expression networks utilising publicly available transcriptomic data from multiple brain regions (provided by the Genotype-Tissue Expression Consortium and UK Brain Expression Consortium), whilst secondary networks were used to examine cell-type specificity. Reverse engineering of gene regulatory networks generated regulons of the complex, which were tested for heritability using stratified linkage disequilibrium score regression and then validated in vitro using the QuantiGene multiplex assay. Significant clustering of non-specific lethal genes was revealed alongside Parkinsons disease-associated genes in frontal cortex primary co-expression modules. Both primary and secondary co-expression modules containing these genes were enriched for mainly neuronal cell types. Regulons of the complex contained Parkinsons disease-associated genes and were enriched for biological pathways genetically linked to disease. When examined in a neuroblastoma cell line, 41% of prioritised gene targets showed significant changes in mRNA expression following KANSL1 or KAT8 perturbation. In conclusion, genes encoding the non-specific lethal complex are highly correlated with and regulate genes associated with Parkinsons disease. Overall, these findings reveal a potentially wider role for this protein complex in regulating genes and pathways implicated in Parkinsons disease.

neuroscience↗

bioRxiv2022

Single-nucleus co-expression networks of dopaminergic neurons support iron accumulation as a plausible explanation to their vulnerability in Parkinson's disease

Motivationgene co-expression networks have been widely applied to identify critical genes and pathways for neurodegenerative diseases such as Parkinsons and Alzheimers 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. ResultsWe 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 Parkinsons 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 implementationThe CoExpNets implementations are available as R packages: scCoExpNets 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 Contactalicia.gomez1@um.es Supplementary informationsupplementary data is available online.

bioinformatics↗

bioRxiv2022

Analysis of nucleus and cytoplasm-specific RNA fractions demonstrates that a significant proportion of the genetic regulation of gene expression across the human brain occurs post-transcriptionally

Gaining insight into the genetic regulation of gene expression in human brain is key to the interpretation of genome-wide association studies for major neurological and neuropsychiatric diseases. Expression quantitative trait loci (eQTL) analyses have largely been used to achieve this, providing valuable insights into the genetic regulation of steady-state RNA in human brain, but not distinguishing between molecular processes regulating transcription and stability. RNA quantification within cellular fractions can disentangle these processes in cell types and tissues which are challenging to model in vitro. We investigated the underlying molecular processes driving the genetic regulation of gene expression specific to a cellular fraction using allele-specific expression (ASE). Applying ASE analysis to genomic and transcriptomic data from paired nuclear and cytoplasmic fractions of anterior prefrontal cortex, cerebellar cortex and putamen tissues from 4 post-mortem neuropathologically-confirmed control human brains, we demonstrate that a significant proportion of genetic regulation of gene expression occurs post-transcriptionally in the cytoplasm, with genes undergoing this form of regulation more likely to be synaptic. These findings have implications for understanding the structure of gene expression regulation in human brain, and importantly the interpretation of rapidly growing single-nucleus brain RNA-sequencing and eQTL datasets, where cytoplasm-specific regulatory events could be missed.

genomics↗

bioRxiv2021

Algorithms for discovery of cis-eQTL signals in woody species: the vine (Vitis vinifera L.) as a study model

Expression quantitative trait loci (eQTLs), are associations between genetic variants, such as Single Nucleotide Polymorphisms (SNPs), and gene expression. eQTLs are an important tool to understand the genetics of gene expression of complex phenotypes. eQTLs analysis are common in human studies and in model species such as mice, rats and yeast but are very scarce in wood crop species such as fruit trees or grapevines. In this study a comprehensive bioinformatic pipeline has been carried out using genomics and expression data from 10 genotypes of grape, which has been used as model species. As a result of this study a total of 10,618 genetic variants that regulate gene expression levels of 525 genes were detected. A 78% of them, 411, received a functional annotation from UniProtKB or DAVID, between the annotated protein-coding genes are Germin-like proteins (GLPs), auxin-regulatory factors, GRFS, ANK_REP_REGION domain-containing protein, Kinesin motor domain-containing protein or RPP13like protein 2(LOC100852873). This new inventory of cis eQTLs influencing gene expression during the ripening of fruits of Vitis vinifera L. will be an important resource for future research to understand the mechanistic basis for variation in gene regulation in this species. In the future, this methodology may be applied to other woody species, once the necessary databases are generated.

plant biology↗

bioRxiv2021

PhenoExam: an R package and Web application for the examination of phenotypes linked to genes and gene sets

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 Parkinsons 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/.

bioinformatics↗

bioRxiv2021

Detection of pathogenic splicing events from RNA-sequencing data using dasper

Although next-generation sequencing technologies have accelerated the discovery of novel gene-to-disease associations, many patients with suspected Mendelian diseases still leave the clinic without a genetic diagnosis. An estimated one third of these patients will have disorders caused by mutations impacting splicing. RNA-sequencing has been shown to be a promising diagnostic tool, however few methods have been developed to integrate RNA-sequencing data into the diagnostic pipeline. Here, we introduce dasper, an R/Bioconductor package that improves upon existing tools for detecting aberrant splicing by using machine learning to incorporate disruptions in exon-exon junction counts as well as coverage. dasper is designed for diagnostics, providing a rank-based report of how aberrant each splicing event looks, as well as including visualization functionality to facilitate interpretation. We validate dasper using 16 patient-derived fibroblast cell lines harbouring pathogenic variants known to impact splicing. We find that dasper is able to detect pathogenic splicing events with greater accuracy than existing LeafCutterMD or z-score approaches. Furthermore, by only applying a broad OMIM gene filter (without any variant-level filters), dasper is able to detect pathogenic splicing events within the top 10 most aberrant identified for each patient. Since using publicly available control data minimises costs associated with incorporating RNA-sequencing into diagnostic pipelines, we also investigate the use of 504 GTEx fibroblast samples as controls. We find that dasper leverages publicly available data effectively, ranking pathogenic splicing events in the top 25. Thus, we believe dasper can increase diagnostic yield for a pathogenic splicing variants and enable the efficient implementation of RNA-sequencing for diagnostics in clinical laboratories.

bioinformatics↗

bioRxiv2021

Genetic variability associated with OAS1 expression in myeloid cells increases the risk of Alzheimer's disease and severe COVID-19 outcomes

Genome-wide association studies of late-onset Alzheimers disease (AD) have highlighted the importance of variants associated with genes expressed by the innate immune system in determining risk for AD. Recently, we and others have shown that genes associated with variants that confer risk for AD are significantly enriched in transcriptional networks expressed by amyloid-responsive microglia. This allowed us to predict new risk genes for AD, including the interferon-responsive oligoadenylate synthetase 1 (OAS1). However, the function of OAS1 within microglia and its genetic pathway are not known. Using genotyping from 1,313 individuals with sporadic AD and 1,234 control individuals, we confirm that the OAS1 variant, rs1131454, is associated with increased risk for AD and decreased OAS1 expression. Moreover, we note that the same locus was recently associated with critical illness in response to COVID-19, linking variants that are associated with AD and a severe response to COVID-19. By analysing single-cell RNA-sequencing (scRNA-seq) data of isolated microglia from APPNL-G-F knock-in and wild-type C57BL/6J mice, we identify a transcriptional network that is significantly upregulated with age and amyloid deposition, and contains the mouse orthologue Oas1a, providing evidence that Oas1a plays an age-dependent function in the innate immune system. We identify a similar interferon-related transcriptional network containing OAS1 by analysing scRNA-seq data from human microglia isolated from individuals with AD. Finally, using human iPSC-derived microglial cells (h-iPSC-Mg), we see that OAS1 is required to limit the pro-inflammatory response of microglia. When stimulated with interferon-gamma (IFN-{gamma}), we note that cells with lower OAS1 expression show an exaggerated pro-inflammatory response, with increased expression and secretion of TNF-. Collectively, our data support a link between genetic risk for AD and susceptibility to critical illness with COVID-19 centred on OAS1 and interferon signalling, a finding with potential implications for future treatments of both AD and COVID-19, and the development of biomarkers to track disease progression.

genetics↗

bioRxiv2021

Leveraging omic features with F3UTER enables identification of unannotated 3'UTRs for synaptic genes

There is growing evidence for the importance of 3 untranslated region (3UTR) dependent regulatory processes. However, our current human 3UTR catalogue is incomplete. Here, we developed a machine learning-based framework, leveraging both genomic and tissue-specific transcriptomic features to predict previously unannotated 3UTRs. We identify unannotated 3UTRs associated with 1,513 genes across 39 human tissues, with the greatest abundance found in brain. These unannotated 3UTRs were significantly enriched for RNA binding protein (RBP) motifs and exhibited high human lineage-specificity. We found that brain-specific unannotated 3UTRs were enriched for the binding motifs of important neuronal RBPs such as TARDBP and RBFOX1, and their associated genes were involved in synaptic function and brain- related disorders. Our data is shared through an online resource F3UTER (https://astx.shinyapps.io/F3UTER/). Overall, our data improves 3UTR annotation and provides novel insights into the mRNA-RBP interactome in the human brain, with implications for our understanding of neurological and neurodevelopmental diseases.

bioinformatics↗

bioRxiv2021

Multi-Modality Machine Learning Predicting Parkinsons Disease

BackgroundPersonalized medicine promises individualized disease prediction and treatment. The convergence of machine learning (ML) and available multi-modal data is key moving forward. We build upon previous work to deliver multi-modal predictions of Parkinsons Disease (PD). MethodsWe performed automated ML on multi-modal data from the Parkinsons Progression Marker Initiative (PPMI). After selecting the best performing algorithm, all PPMI data was used to tune the selected model. The model was validated in the Parkinsons Disease Biomarker Program (PDBP) dataset. Finally, networks were built to identify gene communities specific to PD. FindingsOur initial model showed an area under the curve (AUC) of 89.72% for the diagnosis of PD. The tuned model was then tested for validation on external data (PDBP, AUC 85.03%). Optimizing thresholds for classification, increased the diagnosis prediction accuracy (balanced accuracy) and other metrics. Combining data modalities outperforms the single biomarker paradigm. UPSIT was the largest contributing predictor for the classification of PD. The transcriptomic data was used to construct a network of disease-relevant transcripts. InterpretationWe have built a model using an automated ML pipeline to make improved multi-omic predictions of PD. The model developed improves disease risk prediction, a critical step for better assessment of PD risk. We constructed gene expression networks for the next generation of genomics-derived interventions. Our automated ML approach allows complex predictive models to be reproducible and accessible to the community. FundingNational Institute on Aging, National Institute of Neurological Disorders and Stroke, the Michael J. Fox Foundation, and the Global Parkinsons Genetics Program. RESEARCH IN CONTEXTO_ST_ABSEvidence before this studyC_ST_ABSPrior research into predictors of Parkinsons disease (PD) has either used basic statistical methods to make predictions across data modalities, or they have focused on a single data type or biomarker model. We have done this using an open-source automated machine learning (ML) framework on extensive multi-modal data, which we believe yields robust and reproducible results. We consider this the first true multi-modality ML study of PD risk classification. Added value of this studyWe used a variety of linear, non-linear, kernel, neural networks, and ensemble ML algorithms to generate an accurate classification of both cases and controls in independent datasets using data that is not involved in PD diagnosis itself at study recruitment. The model built in this paper significantly improves upon our previous models that used the entire training dataset in previous work1. Building on this earlier work, we showed that the PD diagnosis can be refined using improved algorithmic classification tools that may yield potential biological insights. We have taken careful consideration to develop and validate this model using public controlled-access datasets and an open-source ML framework to allow for reproducible and transparent results. Implications of all available evidenceTraining, validating, and tuning a diagnostic algorithm for PD will allow us to augment clinical diagnoses or risk assessments with less need for complex and expensive exams. Going forward, these models can be built on remote or asynchronously collected data which may be important in a growing telemedicine paradigm. More refined diagnostics will also increase clinical trial efficiency by potentially refining phenotyping and predicting onset, allowing providers to identify potential cases earlier. Early detection could lead to improved treatment response and higher efficacy. Finally, as part of our workflow, we built new networks representing communities of genes correlated in PD cases in a hypothesis-free manner, showing how new and existing genes may be connected and highlighting therapeutic opportunities.

genomics↗

bioRxiv2020

Modeling multifunctionality of genes with secondary gene co-expression networks in human brain provides novel disease insights

MotivationCo-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 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. ResultsWe 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 by 46.73%. Moreover, GMSCA predicts that the SNCA gene, traditionally associated to work mainly in neurons, also plays a relevant function in oligodendrocytes. AvailabilityThe tool is available at GitHub,https://github.com/drlaguna/GMSCA as open source software. ImplementationGSMCA is implemented in R.

bioinformatics↗

bioRxiv2020

CoExp Web, a web tool for the exploitation of co-expression networks

SummaryThis paper presents the CoExp Web tool. CoExp is a software tool and a Web resource that fosters research on co-expression network models by providing a highly usable means to exploit and share multiple network models. Availability and Implementationthe Web resource is accessible at https://snca.atica.um.es/coexp/. The software at the front and back-ends is accessible at https://github.com/SoniaRuiz/CoExp_Web. Contactskgtsg0@ucl.ac.uk (Sonia Garcia-Ruiz)

bioinformatics↗