During the 2019-2020 academic year, a total of 94 PhD theses were read in the PhD Programme in Biomedicine run by the DCEXS. Their corresponding examination committees put 35 of them forward to compete for the PhD Special Awards. 

The Special Award Comittee that awarded the ten candidates was formed by Robert Castelo, Marc Güell, Cristina López Rodriguez, Elena Bosch, Pilar Navarro, Pilar Rivera, Fernando García Benavides, Joaquim Gea, Mariano Sentí, Núria Centeno, Ruben Vicente. The granted candidates are:

Pablo Baeza Centurión: “Understanding alternative splicing using deep mutagenesis”

Changes in pre-mRNA splicing are hypothesised to be frequent causes of human genetic disease. Splicing can be disrupted by mutations in splice sites themselves but also by mutations elsewhere within exons and introns, including synonymous mutations that do not cause a change in protein sequence. Indeed, recent experiments have shown that mutations throughout exons and introns can cause large changes in splicing, suggesting that changes in splicing may be even more important as potential disease mechanisms than previously suspected.

The thesis by Pablo Baeza Centurión, directed by Dr. Ben Lehner from the Centre for Genomic Regulation, used deep mutagenesis (in which many mutations are inserted into a genomic region of interest) to understand how mutations combine to affect splicing, coupled with a mathematical model of the splicing reaction. The results showed that the mathematical model could recreate the observations of the deep mutagenesis experiment, suggesting that there is a scaling law for the effects of mutations on alternative splicing and provide a rare example of a global predictive law in biology. Further experiments also showed that using their model it is quite straightforward to predict genome-wide the exons that are most likely to have their splicing altered by mutations and therefore, it provides a useful and practical framework for predicting when such mutations are likely, or not, to contribute to disease.

Selected publications from the thesis:

• Baeza-Centurion* P, Miñana* B, Schmiedel JM, Valcárcel** J, Lehner** B. Combinatorial genetics reveals a scaling law for the effects of mutations on splicing. Cell, 173(3):549-563, 2019.
• Baeza-Centurion* P, Miñana* B, Valcárcel** J, Lehner** B. Mutations primarily alter the inclusion of alternatively-spliced exons. eLife, 9:e59959, 2020.

Júlia Domingo Espinós: “Using deep mutagenesis to understand genetic and physical interactions”

A mutation will not always have the same consequences in an individual. One of the main reasons why this occurs is the presence of additional genetic variation. The dependency of the effects of mutations on the genetic background, also termed genetic interaction (or epistasis), has been extensively characterized between pairs of mutations within and between genes. However, the pairwise mapping of interactions only provides a limited view of the whole genotype space, which has a vast combinatorial size.

The thesis by Júlia Domingo Espinós, directed by Dr. Ben Lehner from the Centre for Genomic Regulation, introduced a large combination of 14 mutations (covering single, double and higher order combinations) to a conditionally essential gene, a tRNA, and precisely quantified their effects on cellular fitness in a single experiment. This experiment measured accurately for the first time interactions beyond pairs of mutations and estimated the contribution of interactions at multiple orders to predict phenotype from genotype. The analysis of those contributions suggest that accurate genetic predictions require mutation effects to be measured across different genetic backgrounds and the use of sparse higher order interactions.

Selected publications from the thesis:

• Domingo J, Diss G, Lehner B. Pair wise and higher-order genetic interactions during the evolution of a trNA. Nature, 558:117–121, 2018. 
• Domingo* J, Baeza-Centurion* P, Lehner B. The causes and consequences of genetic interactions (Epistasis), Annual review of genomics and human genetics, 20:433-460, 2019.

Joan Frigola Rissech: “Determinants of the local mutation rate variability along the genome”

Our cells are constantly exposed to different types of agents and processes that may damage their DNA. To deal with these lesions, they are equipped with a collection of repair mechanisms. However, a fraction of these lesions may not be repaired and could lead to the appearance of mutations after replication. DNA mutations are permanent changes in the DNA sequence which, when affecting DNA regions that contain information relevant for the cells proper functioning, may have severe consequences such as cell death or disease, including tumor formation.

The thesis by Joan Frigola Rissech, directed by Dr. Abel González-Perez and Dr. Nuria Lopez-Bigas from the Institute for Research in Biomedicine, studied mutational processes and how DNA mutations are distributed along the genome, taking into consideration the different damaging processes cells have been exposed to. The findings show that there is a specific DNA repair mechanism called mismatch repair pathway that preferentially repairs the most important regions of the genes (exons). This may have relevant implications since several statistical methods used in cancer genomics assume the mutation rate to be constant along the genes. Additionally, they also show that mutational hot-spots in the promoter regions of UV exposed cells are the result of an impaired DNA damage repair activity and an increased damage formation. These observations may help to better understand tumor formation and evolution, and teaches us about basic biological processes related to the DNA damage formation and repair.

Selected publications from the thesis:

• Frigola* J, Sabarinathan* R, Mularoni L, Gonzalez-Perez A and Lopez-Bigas N. Reduced mutation rate in exons due to differential mismatch repair. Nature Genetics, 49:1684-1692, 2017. 
• Frigola* J, Sabarinathan R, Gonzalez-Perez A and Lopez-Bigas N. Variable interplay of UV-induced DNA damage and repair at transcription factor binding sites. Nucleic Acids Research, 49:891-901, 2021.

Diego Garrido Martin: “A multivariate approach to study the genetic determinants of phenotypic traits”

Alternative splicing (AS) is the process through which multiple transcripts are produced from a single gene. Transcriptome profiling of large cohorts of genotyped individuals allows to identify genetic variants affecting AS (splicing quantitative trait loci or sQTLs). sQTL analyses provide valuable insights into the mechanisms underlying complex traits and diseases. To date, most methods used for sQTL mapping assess association between genetic variants and the abundance of individual transcripts, or the splicing of individual exons or introns. However, this ignores the strongly correlated structure of AS measurements.

The thesis by Diego Garrido Martin, directed by Dr. Roderic Guigó and Dr. Miquel Calvo Llorca, from the Centre for Genomic Regulation, developed an approach that takes into account the intrinsically multivariate nature of AS, testing genetic variants for association with a vector of AS phenotypes (e.g. a gene’s relative isoform abundances). They employ it to study the multi-tissue transcriptome GTEx dataset, generating a comprehensive catalogue of sQTLs in the human genome. The analysis of this catalogue contributed to understand the mechanisms of AS regulation.

Selected publications from the thesis:

• Garrido-Martín* D, Palumbo E, Guigó R and Breschi A. ggsashimi: Sashimi plot revised for browser- and annotation-independent splicing visualization. PLOS Computational Biology. 14:e1006360, 2018.
• Garrido-Martín D, Borsari B, Calvo M, Reverter F and Guigó R. Identification and analysis of splicing quantitative trait loci across multiple tissues in the human genome. Nature Communications, 12:727, 2021.

José Jiménez Luna: “Machine learning in structural biology and chemoinformatics”

Deep learning approaches have become increasingly popular in the last few years thanks to their state-of-the-art performance in fields such as computer vision and natural language understanding.

The thesis by José Jiménez Luna, directed by Dr. Gianni de Fabritiis adapted such approaches to the domains of structural biology and chemoinformatics. This was done by the development of a novel three-dimensional biomolecular representation of proteins and small molecules that can be used in conjunction with 3D-convolutional neural networks for a variety of tasks. They tested the applicability of such approaches in several critical problems in the early pipeline of drug discovery, such as protein binding site prediction, structure-based affinity prediction, target selectivity elucidation and de novo drug design of novel chemical entities. Their results showed that the proposed array of approaches developed throughout the thesis are competitive and, in many cases, achieved increased performance over existing approaches, validating them with pharmaceutical commercial partners such as Pfizer, Biogen, Janssen and Novartis. The software has been also made available to the public at https://playmolecule.org. 

Selected publications from the thesis:

• Jiménez J, Doerr S, Martínez-Rosell G, Rose AS and De Fabritiis G. DeepSite: Protein-binding site predictor using 3D-convolutional neural networks. Bioinformatics, 33:3036-3042, 2017.
• Jiménez J, Skalic M, Martínez-Rosell G and De Fabritiis G. Kdeep: protein–ligand absolute binding affinity prediction via 3D-convolutional neural networks. Journal of Chemical Information and Modeling, 58:287-296, 2018.

Oriol Llorà Batlle: “Characterization of sexual commitment and the early steps of sexual development in the human malaria parasite Plasmodium falciparum”

Malaria is a vector-borne disease that affects 200 million people and causes 445.000  deaths every year mainly due to the parasite Plasmodium falciparum, responsible for the most severe form of malaria. Humans acquire malaria by the bite of an infected female Anopheles mosquito. Once in the human host, parasites establish a 48-hours chronic cycle of asexual growth inside erythrocytes. At each cycle, a small proportion of the parasites differentiate into male or female gamete precursors called gametocytes, which are the only stage capable of infecting the mosquito vector. Therefore, gametocytes are an attractive stage to target – preventing transmission is considered essential for malaria eradication. 

The thesis by Oriol Llorà Batlle, directed by Dr. Alfred Cortés Closas from the Barcelona Institute for Global Health, identified a new route of sexual differentiation that challenges the current textbook view of the transmission of the parasite and highlights its plasticity in this key step of the life cycle. Using a novel combination of conditional recombinases and CRISPR/Cas9 technology, they also built a tool for the study of sexual differentiation, potentially contributing towards the development of transmission-blocking strategies.

Selected publications from the thesis:

• Bancells, C, Llorà-Batlle O., Poran A, Nötzel C, Rovira-Graells N, Elemento O, Kafsack BFC and Cortés, A. Revisiting the initial steps of sexual development in the malaria parasite Plasmodium falciparum. Nature Microbiology, 4:144-154, 2019.
• Llorà-Batlle O, Michel-Todó L, Witmer K, Toda H, Fernández-Becerra C, Baum J and Cortés A. Conditional expression of PfAP2-G for controlled massive sexual conversion in Plasmodium falciparum. Science Advances, 6:eaaz5057, 2020.

Lídia Mateo: “A Computational Toolkit to Boost Precision Oncology with Pharmacogenomics”

Molecular profiling of personal cancer genomes, and the identification of actionable vulnerabilities and drug-response biomarkers, are the basis of precision oncology. Cancer is a complex disease and, as such, the future of clinical research needs to go beyond single-gene biomarkers in order to expand the applicability domain of targeted therapies and maximize the benefit of patients who receive them.

The thesis by Lídia Mateo, directed by Dr. Patrick Aloy
from the Institute for Research in Biomedicine, developed a computational strategy to tailor a cancer patient’s treatment to the specific molecular alterations of their tumor (Targeted Cancer Therapy for You or TCT4U). Using this software, they uncovered and exploited driver alteration co-occurrence patterns from patient-derived mouse xenografts (PDXs) and, for each treatment, compared the mutational profiles of sensitive and resistant PDXs, and developed a methodology that prioritizes the best possible treatment for each tumor based on its oncogenomic profile. Furthermore, to aid in the drug-molecular profile matching process, they also developed two accompanying tools that enable the contextualization and visualization of the molecular portraits of individual patient tumors (Cancer PanorOmics) and cohorts of patients (Oncogenomic Landscapes). This global molecular view of individual tumors and patient cohorts, together with the appropriate genetic counseling and medical advice, can contribute to the identification of actionable alterations ultimately guiding the clinical decision-making process. Altogether, this toolkit will contribute to expand the applicability domain of precision oncology.

Selected publications from the thesis:

• Mateo L, Guitart-Pla O, Duran-Frigola M and Aloy P. Exploring the OncoGenomic Landscape of cancer. Genome Medicine, 10:61, 2018.
• Mateo L, Duran-Frigola M, Gris-Oliver A, Palafox M, Scaltrity M, Razavi P, Chandarlapaty S, Arribas J, Bellet M, Serra V and Aloy P. Personalized cancer therapy prioritization based on driver alteration co-occurrence patterns. Genome Medicine, 12:78, 2020.

María Gabriela Prado Peralta: “Understanding the role of body weight and composition on lung function growth and decline”

Lifetime lung function is related with quality of life and all-cause mortality in the general population. Considering the steadily increase of mean body weight and the pandemic levels of overweight and obesity around the globe, several studies have assessed the effects of body weight on lung function. There is evidence showing that post-natal weight growth characteristic can affect lung function in childhood. Childhood overweight and obesity have also been related to lung function levels, but the reported associations remain inconsistent. In addition, overweight and obesity, as well as excessive weight gain, during adult life have been suggested to increase the risk of accelerated lung function decline. 

The thesis by María Gabriela Prado Peralta, directed by Dr. Judith Garcia Aymerich from the Barcelona Institute for Global Health, aimed to assess the associations of body weight and composition with lung function growth and decline, overcoming some of the limitations of previous research, such as using body mass index, which does not allow one to distinguish the different effects of lean body mass and fat mass on lung function. They confirmed the deleterious effect of obesity on lung function over lifespan and provided novel and relevant contributions for the respiratory field. They also showed, for the first time, that body composition trajectories from childhood to adolescence are associated with lung function growth and that lean body mass and fat mass have opposite effects on lung function during childhood and adolescence. These findings highlight the importance of assessing body composition when studying the health effects of body weight on respiratory health and suggest that public health policies and clinical interventions aiming to reduce respiratory morbidity, should target body composition in addition of body weight. Another key finding of this thesis is showing that the effects of body weight on lung function are reversible both in early childhood and in adulthood. 

Selected publications from the thesis:

• Peralta GP, Fuertes E, Granell R, Mahmoud O, Roda C, Serra I, Jarvis D, Henderson J, Garcia-Aymerich J. Childhood body composition trajectories and  adolescent lung function: Findings from the ALSPAC study. American Journal of Respiratory and Critical Care Medicine, 200:75-83, 2019.
• Peralta GP, Abellan A, Montazeri P, Basterrechea M, Esplugues A, González S, Roda C, Santa Marina L, Sunyer J, Vrijheid M, Casas M, Garcia-Aymerich J. Early childhood growth is associated with lung function at seven years: a prospective population-based study. European Respiratory Journal, 56:2000157, 2020.

Marc Talló Parra: “Circular RNAs: from host RNA molecules to novel broad-spectrum antivirals”

Circular RNAs (circRNAs) are highly stable single-stranded covalently closed RNA molecules generated by an alternative splicing mechanism called back-splicing. During decades they were considered splicing artifacts with no functional value. This changed a few years ago when one circRNA was described to be crucial in brain development. The role of circRNAs on infections by RNA viruses was mostly unexplored.

The thesis by Marc Talló Parra, directed by Dr. Juana Díez from the Universitat Pompeu Fabra, uncovered a functional role of natural host circRNAs in viral RNA infection and exploited the high stability of circRNA molecules to generate a novel antiviral platform that allows ultrafast responses against future viral outbreaks and pandemics. They showed that infections by RNA viruses have a profound impact on host circRNA landscapes in the infected cells. Using hepatitis C virus as a model system, they found that some of the virus-induced up-regulated circRNAs have a pro-viral effect by binding and thus sequestering host proteins with antiviral effects. These results uncovered a complex novel layer of interactions between RNA viruses and the host-encoded circRNAs. Because of their long half-lives, circRNAs will likely play important roles in viral pathogenesis.

Pandemics caused by yet unknown viruses that infect and spread within humans for the very first time, such as the present SARS-CoV-2, are recurrent and will continue to threaten mankind. Their rapid spread is favoured by high population densities within cities and massive traveling activities. Thus, a fast development of an antiviral drug against any new viral pathogen is key to timely respond to novel viral outbreaks and to contain potential pandemics. Part of the research conducted in this thesis built a novel RNA-based drug platform that enables the generation of novel candidate antiviral molecules within 2 to 4 weeks after the sequence of a viral genome is disclosed. This drug platform is based on stable artificial RNA molecules that can be designed to target essential viral genome structures and efficiently inactivate virus replication. As a proof of concept, the thesis successfully validated RNA molecules that inhibit HCV, DENV, CHIKV or WNV. Furthermore, they also generated RNAs with broad-spectrum antiviral capacity. This work is not yet published because it forms part of a patent application.

Selected publications from the thesis:

• Chen* TC, Tallo-Parra* M, Cao QM, Kadener S, Böttcher R, Pérez-Vilaró G, Boonchuen P, Somboonwiwat K, Díez J and Sarnow P. Host-derived circular RNAs display proviral activities in Hepatitis C virus-infected cells. PLOS Pathogens, 16:1-23, 2020.
• Diez J, Tallo-Parra M, Dotu I. Artificial circular RNAs for treating viral infections European Patent Application 20 382 569.0. Titular Entity: Universitat Pompeu Fabra. Priority country: Spain. Priority date: 31/12/18.

Antonio Torres Méndez: “Origin and evolution of neural microexons”

Post-transcriptional gene regulatory programs control multiple aspects of neuronal biology, and their dysregulation has been associated with many human neurological disorders. Alternatively spliced neural-enriched microexons, are very short exons that are activated during neuronal differentiation and generate protein isoforms with one to nine additional amino acids. Their discovery was surprising and mechanistically challenging, since their short length was expected to preclude standard interactions needed for exon definition. Interestingly, they represent the most conserved type of alternative splicing in vertebrates and are misregulated in some human patients with autism spectrum disorder.

The thesis by Antonio Torres Méndez, directed by Dr. Manuel Irimia Martínez from the Centre for Genomic Regulation, studied the post-transcriptional regulation of neural microexons to address fundamental questions on how gene regulatory programs can originate and evolve. First, by searching for neural microexons in non-vertebrate animals using transcriptomics, they found that neural microexons predate the last common ancestor of bilaterians (over 600 million years ago) and share similar sequence regulatory features across species. Second, by generating transgenic flies they found widespread neurological alterations in locomotion, ageing, sleep and metabolism, among others. In collaboration with neuroscientists, they were able to show that these defects stem from altered neuronal activity rather than changes in neuroanatomy.

Selected publications from the thesis:

• Torres-Méndez* A, Bonnal* S, Marquez Y, Roth J, Iglesias M, Permanyer J, Almudí I, O’Hanlon D, Guitart T, Soler M, Gingaras A-C, Gebauer F, Rentzsch F, Blencowe BJ, Valcárcel J and Irimia M. A novel protein domain in an ancestral splicing factor drove the evolution of neural microexons. Nature Ecology and Evolution, 3:619-701, 2019.

 More information