The group’s main objective is to contribute to a more predictive, preventive, and personalized medicine through the integration of omics, clinical, and population-level data. To achieve this, we develop models, algorithms, computational resources, and validation strategies aimed at facilitating the use of complex biomedical data in real-world research and healthcare settings.

The group has a strong translational and collaborative focus, with expertise in clinical genomics, rare diseases, cancer, molecular epidemiology, drug repurposing, real-world data analysis, and artificial intelligence applied to healthcare.

Research Areas

• Clinical Genomics

The group develops and applies bioinformatics methods for the interpretation of genomic data in clinical and translational settings. This research area includes the analysis of genetic variants, the integration of genomic and clinical information, the prioritization of disease-associated genes and variants, and the development of resources that facilitate the incorporation of genomics into clinical practice.

Our activity in this field is closely linked to participation in CIBERER (https://www.ciberer.es/) and builds on a long-standing track record in rare diseases, hereditary cancer, and personalized medicine. The group has contributed to the development of resources and strategies for the interpretation of individual and population genomes, including the Spanish Variant Server (CSVS; https://csvs.babelomics.org/) and SPACNACS (https://csvs.clinbioinfosspa.es/spacnacs/), as well as genomic surveillance and molecular epidemiology initiatives.

This area also encompasses cancer research, genetic susceptibility studies, germline and somatic variant analyses, and molecular epidemiology projects such as SIEGA (https://www.clinbioinfosspa.es/projects/siega/), which focus on the use of genomic sequencing for the surveillance, characterization, and monitoring of infectious agents and their impact on public health. This work embraces a One Health perspective.

• Systems Medicine

Systems medicine is one of the group’s core research areas. Our goal is to understand disease as the result of coordinated alterations in molecular networks, signaling pathways, cellular processes, and complex pathophysiological mechanisms.

In this context, the group has developed and applied methodologies such as HiPathia (http://hipathia.babelomics.org/), designed to model the functional activity of signaling pathways from omics data. These models enable the interpretation of transcriptomic and genomic data from a mechanistic perspective, the identification of biological processes altered in disease, and the generation of hypotheses regarding mechanisms of action, biomarkers, and potential therapeutic interventions.

This research area also includes mechanism-based drug repurposing, disease modeling, disease map analysis, and hypothesis validation using real-world clinical data. The objective is to connect molecular information with clinical phenotypes and health outcomes, facilitating the translation of computational biology into biomedical and clinical applications.

• Real-World Evidence Generation

The group develops methods for generating evidence from real-world data, including electronic health records, population registries, administrative databases, pharmacy records, laboratory results, medical imaging, and other data generated during routine clinical care.

This line of research is based on the premise that routinely collected clinical data constitute a strategic source of knowledge for biomedical research, outcomes evaluation, healthcare planning, and personalized medicine. The rigorous, ethical, secure, and reproducible use of these data enables the study of entire populations, the identification of disease patterns, the evaluation of clinical trajectories, and the development of predictive models applicable in real-world settings.

Research topics include the development of early disease predictors, risk stratification models, retrospective population studies, health outcomes analyses, treatment evaluation, and the external validation of clinical algorithms. Through its participation in initiatives such as IMPaCT Data (https://impact-data.bsc.es/), which the group co-leads, and OmicsSpace (https://omicspace.iislafe.es/), the group has a particular interest in federated and standardized approaches that enable the reuse of clinical data while preserving privacy, governance, and analytical traceability.

• Artificial Intelligence in Medicine

The group develops and applies advanced artificial intelligence techniques for the analysis of complex biomedical data. This research area includes interpretable machine learning, clinical predictive models, multimodal data integration, generative artificial intelligence, biomedical foundation models, and AI-based systems that support research activities.

A specific area of interest is the generation of synthetic patient data, conceived as a tool to facilitate research, model training and validation, methodological evaluation, and the secure sharing of information when access to real-world data is restricted. The group addresses both the technical aspects and the ethical, methodological, and regulatory challenges associated with the generation and use of synthetic health data.

La entrada Artificial Intelligence and Data in Health se publicó primero en GRIB.

Main Research lines:

Highlighted publications:

La entrada Computational Epigenomics se publicó primero en GRIB.

Main research lines:

• The neuroactivity of the gut microbiome

  Alterations in the composition of the gut microbiome have been associated with multiple neuropathologies. The members of the microbiome metabolise a myriad of neuroactive compounds, including GABA, serotonin, or dopamine which, through the microbiome-gut-brain axis, could influence mental health. We develop computational tools to profile the synthesis and degradation of such compounds in prokaryotic genomes and metagenomes, to better understand how alterations in the gut microbiome could be linked with mental health related conditions.

• The social transmission of the microbiome

  Our microbiome is seeded during birth, and then shaped by transmission from proximate individuals. We use strain-level resolution metagenomics profiling to track microbiome transmission through social interaction, and how this shapes the microbiome in health and disease.

Highlighted publications

-Valles-Colomer M, Blanco-Miguez A, Manghi P, Asnicar F, Dubois L, Golzato D, Armanini F, Cumbo F, Huang KD, Manara S, Masetti G, Pinto F, Piperni E, Puncochar M, Ricci L, Zolfo M, Farrant O, Goncalves A, Selma-Royo M, Binetti AG, Becerra JE, Han B, Lusingu J, Amuasi J, Amoroso L, Visconti A, Steves CM, Falchi M, Filosi M, Tett A, Last A, Xu Q, Qin N, Qin H, May J, Eibach D, Corrias MV, Ponzoni M, Pasolli E, Spector TD, Domenici E, Collado MC, Segata N. The person-to-person transmission landscape of the gut and oral microbiomes. Nature 2023; 614(7946): 125-135.

-Valles-Colomer M, Bacigalupe R, Vieira-Silva S, Suzuki S, Darzi Y, Tito RY, Yamada T, Segata N, Raes J, Falony G. Variation and transmission of the human gut microbiota across multiple familial generations. Nature Microbiology 2022; 7(1): 87-96.

-Valles-Colomer M, Falony G, Darzi Y, Tigchelaar EF, Wang J, Tito RY, Schiweck C, Kurilshikov A, Joossens M, Wijmenga C, Claes S, Van Oudenhove L, Zhernakova A, Vieira-Silva S, Raes J. The neuroactive potential of the human gut microbiota in quality of life and depression. Nature Microbiology 2019; 4(4): 623-632.

Funding:

-Knowledge Generation Project (PID2022-139328OA-I00) from the Spanish Ministry of Science and Innovation

-La Caixa Incoming Junior Leader (fellowship 132107)

-Beatriz Galindo Junior fellowship (BG22/00172) from the Spanish Ministry of Universities

-EvoMG: Joint Program on Evolutionary Medical Genomics

The web page of the group is: https://www.upf.edu/web/microbiome

La entrada Microbiome Research se publicó primero en GRIB.

In particular, a significant degree of structural variation, including hundreds of copy number variants (insertions, duplications and deletions) and inversions, has been discovered in multiple organisms. In addition, we now have the information on the variation in the expression levels of thousands of genes in diverse tissues and individuals of many species. However, we still know very little about the functional and evolutionary impact of these changes.

Our main line of research therefore deals with the global analysis of polymorphic inversions and other complex regions in the human genome, which are in general poorly characterized and could have unique effects compared to other variants. This ranges from the development of new methods for inversion study and the first database of human polymorphic inversions, to the characterization of their population distribution, functional consequences and selection signatures, as a way to ultimately determine their contribution to complex traits.

La entrada Comparative and functional genomics se publicó primero en GRIB.

• Fold prediction
• Protein docking
• Function prediction and annotation
• Description of the dynamical properties of biomolecules

Research Outline

Protein-protein interactions play a relevant role among the different functions of a cell. Identifying the protein-protein interaction network of a given organism (interactome) is useful to shed light on the key molecular mechanisms within a biological system. A paradox in protein-protein binding is to explain how the unbound proteins of a binary complex recognize each other among a large population within a cell and how they find their best docking interface in a short time-scale. We interrogate protein structure to unveil its function, generate the network of interactions and to relate genes/proteins with diseases by means of exploiting the topology of the network.

Current Projects/Research Lines

On the study of the relationship between sequence, structure and function of proteins: Characterization of the structural motifs involved in the function and interactions between proteins. Development of statistical potentials and analysis of physico-chemical potentials helping to describe the fold and function of proteins and its interactions with other macro-molecules.

On the prediction of protein-protein and protein-DNA interactions: Structural analysis of docking approaches and development of new techniques towards the prediction of binding sites and the mechanisms of interface selection of protein-protein and protein-DNA interactions.

On the analysis of protein interaction networks and its use on bio-medicine, helping to detect potential targets and prioritization of candidate disease-genes. Development of methods to study and integrate information for different types of networks and application on the study of metastasis. Prediction of signalling networks, such as the phosphorylation network and other post-transcriptional modifications, and integration with genomic data, such as microarrays.

Website of group: https://sbi.upf.edu/web/index.php

La entrada Structural Bioinformatics se publicó primero en GRIB.

Nowadays, computational methodologies are widely used in pharmaceutical research. From the initial stages of target validation to safety studies, in silico methods are an integral part of the R&D. Unfortunately, in most cases, the usefulness of these methods is limited by the large complexity of the biological subjects and the multiple biological scales involved in observable outcomes.

In this context, the PharmacoInformatics laboratory aims to improve the current state-of-the-art by developing computational tools based on multi-scale representations of phenomena that go beyond the classic reductionist approaches. We also implement these methodologies in user-friendly software suitable for academic and corporate environments.

Manuel Pastor is the group leader.

Projects

• The H2020 project RISK-HUNT 3R:  RISK assessment of chemicals integrating HUman centric Next generation Testing strategies promoting the 3Rs Activity. RISK-HUNT3R builds on the outcomes of the Horizon 2020 toxicological flagship project EU-ToxRisk, ended by the end of 2021.
• The IMI project eTRANSAFE: Enhancing Translational Safety Assessment through Integrative Knowledge Management, aims to develop an advanced data integration infrastructure together with innovative computational methods to improve the security in drug development process.
• The EU-ToxRisk project, that intends to become the European flagship for animal-free chemical safety assessment. The project will integrate advancements in cell biology, omics technologies, systems biology and computational modelling to define the complex chains of events that link chemical exposure to toxic outcome.
• The iPiE project, which goal is to develop a framework that will provide methodologies to prioritise new and existing medicinal compounds for a comprehensive environmental risk assessment.

Software developed

• FLAME: an open source framework for model development, hosting, and usage in production environments.
• eTOXlab is a flexible modeling framework. It was developed for supporting models predicting the biological properties of chemical compounds (e.g. QSAR models) in production environments. It is distributed embedded within a virtual machine containing all required software for building models and using them for prediction.
• ADAN: Applicability Domain ANalysis, is a tool for assessing the reliability of drug property predictions obtained by in silico methods.

The members of the group belong to UPF .

The web page of the group is https://phi.upf.edu

La entrada PharmacoInformatics se publicó primero en GRIB.

• Systems medicine to identify disease patterns from scientific evidence by applying data mining and network modelling approaches.
• Systems toxicology to reveal molecular mechanisms underlying the toxicity of drugs and other chemicals and to develop predictive machine-learning models.
• Management and exploitation of real-world clinical data.
• Natural language processing to extract relevant information from free-text documents, including scientific publications, clinical records, and social media.
• Development of knowledge management resources in the areas of disease genomics and pharmaceutical R&D.

One of the key developments of the group has been the design and development of the DisGeNET knowledge base of gene-disease associations, which has become a world-wide recognized resource with thousands of users monthly and more than 4,000 bibliographic citations.

The group has a wide record of leadership and participation in international projects. In particular, it is currently coordinating a large-scale scientific initiative, the IMI project “eTRANSAFE: Enhancing Translational Safety Assessment through Integrative Knowledge Management”, which aims to develop an advanced data sharing and integration infrastructure together with innovative software tools to improve the safety assessment during the drug development process.

The group also maintains strong collaborations with the Parc de Salut Mar around clinical data reuse for research purposes, jointly participating in international initiatives in the field, such us the European Health Data Evidence Network or the EMA-funded DARWIN initiative.

Several scientists of the group have founded in 2020 the spin-off company MedBioinformatics Solutions SL, which is focused on providing computational solutions supporting the precision medicine and the pharmaceutical R&D. The company is currently consolidated with physical facilities and several full-time employees. The company shows a positive record of new products, commercial revenues, and participation in EU-funded projects.

The group is also partner in other EU-funded projects, like:

• ERAPerMed PROMPT, Toward PRecisiOn Medicine for the Prediction of Treatment response in major depressive disorder through stratification of combined clinical and -omics phenotypes.
• RISK-HUNT3R, RISK assessment of chemicals integrating HUman centric Next generation Testing strategies promoting the 3Rs.
• Disc4All, an european innovative training networks (ITN) which aims to train young researchers in the tools available to analyze genomic data (DisGeNET) and comorbidities (comoRbidity).

Other examples of recent european projects participated or coordinated by the IBI group are TransQST, FAIRplus, EU-ToxRisk, EHDEN, iPiE, EMIF, MedBioinformatics, ELIXIR-EXCELERATE, eTOX, OpenPHACTS, INBIOMEDvision, VPH NOE, EU-ADR, @neurIST and CancerGRID.

The members of the IBI group belongs to HMRIB and UPF

La entrada Integrative Biomedical Informatics se publicó primero en GRIB.

• In-silico Multi-Receptor Profiling of Antipsychotic Drugs.

The clinical efficacy of antipsychotic drugs has been ascribed to a complex interplay consisting of multiple targets (particularly G protein-coupled receptors, GPCRs) and multiple mechanisms (e.g. signaling bias, receptor crosstalk) which together determine a distinct pattern of cell signaling. In this context, our group focuses on deciphering molecular mechanism of this complex interplay of current antipsychotic drugs that are responsible for their clinical efficacies. Finally, we aim on the discovery of novel pharmacological interventions and the design of new lead structures for antipsychotic therapy.

• GPCR Dimers as Drug Target for the Treatment of Schizophrenia.

Recent studies have described receptor dimerization as relevant mechanism for signaling mediated by GPCRs. We study this phenomenon in the context of antipsychotic drug action. Our group provides support for the design of bivalent ligands that selectively target a specific GPCR dimer. On one hand, such molecular probes are extremely useful to interrogate the contribution of dimers to psychotic conditions. On the other hand, such probes allow us validating the usefulness of GPCR dimers as drug target for the treatment of schizophrenia. Once a target is validated, we apply diverse computational tools to obtain first small drug like molecules (structure-based/ligand-based) towards this target.

• Membrane Lipid-Mediated Effects on GPCR signaling.

The lipid composition of cell membranes can modulate the function of key membrane proteins such as G protein-coupled receptors (GPCRs). As several diseases have suggested to alter relevant biophysical properties of membranes such as density or fluidity, it has become a research priority to understanding the role of membrane environment on the dynamics and function of GPCRs. We address this important question by studying direct and indirect membrane effects on receptor monomers and dimers using all-atom as well as coarse grained simulation setups.

• Database for GPCR dynamics.

The main mission of this project is to provide dynamic insights into crystallized receptors at a public-accessible platform. Main features include deposition /storage of MD data, query tools, visualization of MDs for relevant features as well as basic analysis tools. Importantly, the database is designed to allow not only MD experts but also chemoinformaticians or medicinal chemists to easily browse and use dynamic information on GPCRs. GPCRmd will be integrated into the established and well-known GPCRdb database (http://gpcrdb.org/). This integration will ease the process of gaining popularity and accelerating the deposition of new MD simulations and can make an important contribution to the finding of new drugs towards this important target class. More info here.

Website of group: http://jana-selent.org/

La entrada GPCR Drug Discovery se publicó primero en GRIB.

Research lines

• Reverse engineering the genotype-phenotype map

Genes and molecules are activated in a coordinated manner under finely tuned regulatory programs. High-throughput genetics and genomics data offer a unique opportunity to witness this phenomenon by monitoring the simultaneous action of thousands of genes and millions of genotypes. We try to embrace this complexity by developing computational tools that enable estimating multivariate statistical models from these data, which have the potential to disentangle direct from indirect or spurious effects.

• Variant annotation and filtration

The increased number of individuals profiled by genomics technologies steadily uncovers an increasing number of cases in which pathogenic mechanisms work conditionally on the cellular context where genetic alterations take place, hindering the interpretation of individual mutations. In collaboration with clinical geneticists, we are trying to approach this problem by developing novel methodologies for the annotation and filtration of genetic variants.

• Prematurity and fetal immunity

Intrauterine inflammation and infection increase the risk for perinatal mortality and morbidity and its frequency increases with lower gestational age at birth. In collaboration with pediatricians and obstetricians, we study the transcriptome and proteome of extremely preterm newborns (< 28 weeks of gestational age) to try to understand the extent of molecular changes that participate in the fetal inflammatory response to an intrauterine infection and how these changes lead to adverse neonatal outcome.

Website of group: https://functionalgenomics.upf.edu/index.html

La entrada Functional Genomics se publicó primero en GRIB.

Main Research lines

1. De novo gene birth

The group is performing research on the mechanisms of formation of new genes, with a special emphasis on de novo gene birth. We have shown that proteins that have evolved de novo from previously non-coding parts of the genome tend to be small and positively charged, and that many of them are species-specific. We are investigating the process of the de novo gene birth at very short time scales using population data.

2. Identification of novel classes of tumor-specific antigens

We are investigating gene and protein expression in cancer cells with the objective of identifying novel classes of tumor-specific antigens that may trigger an immune response against the tumor, and which could be useful to develop new anti-cancer vaccines. For this we combine information from transcriptomics, ribosome profiling and immunopeptidomics.

Highlighted publications

Montañés, J.C., Huertas, M., Messeguer, X., Albà, M.M. (2023). Evolutionary trajectories of new duplicated and putative de novo genes. Molecular Biology and Evolution, 40(5):msad098.

William R. Blevins, Jorge Ruiz-Orera, Xavier Messeguer, Bernat Blasco-Moreno, José Luis Villanueva-Cañas, Lorena Espinar, Juana Díez, Lucas B. Carey, M.Mar Albà (2021) Uncovering de novo gene birth in yeast using deep transcriptomics. Nature Communications 12:604.

Group website

For more details please go to http://evolutionarygenomics.upf.edu/

La entrada Evolutionary genomics se publicó primero en GRIB.