Go to the content (press return)

Statistical methods for clinical trials, complex censoring schemes and integrative omics data analysis

Total activity: 37
Type of activity
Competitive project
Funding entity
Funding entity code
255.431,00 €
Start date
End date
The aim of this project is to contribute to the advance on statistical methodologies in survival ana1ysis, clinical trials and integrative analysis of omics data. These three main areas are well related to each other. they not only share a common methodological background but
coincide, often, in the area where they are applied. While omics technologies are the basis for biomarker díscovery and validation, clinical trials are used to investigate the effect of drugs whose action can be rneasured by the previously derived biomarkers and sorne of the
methods in survival analysis form par1 of the core methods to analyze clinical trials.
With respect to ciinical trials, this subproject will consolidate a theory based on the relative efficiency to decide on the best primary
variable. This methodology will be used to address the computation of the sample size when the proportionality of the risks functions is not met. An online platform will be developed to guide clinicians in the decision. A very importan! issue for the pharmaceutical industry is
planning a bioequlvalence trial; new proposals in this project are addressed.
Methodological advances for discrete survival data and with interval-censored data to extend the existing class of hypothesis tests and
providlng new goodness-of-fit methods will be sought. Their application to veterinary studies and to predict the shelf-life of foods will be an outcome of interest. Advances to integrate multistate models, landmark analysis and to incorporate special features of the data such as interval censoring and competing rsks are part of the innovations that we look for.
Deriving biomarkers with high sensitivity and high specificity is a difficult endeavor. To increase the information on which they are built
combining different types of data is a task addressed in thls subproject. We will approach differenl multivariate methods for this type of
integration. Dimension reduction methods applied simultaneously to several, probably incomplete, dataseis will be obtained. Kernel
methods for variable representation in reduced dimension will also be derived. Biological annotatlons will be taken as the basis for
matching datasets with common and non-common individuals so that "virtually complete" dataseis can be obtained to apply the analyses on them. This will improve the possibility for analyzing complex datasets merging different types of omics data. accounting for their
biological interpretabllity -through the use of annotations-and allowing to obtain more sensible biomarkers. System-biology studies where the different levels on which biological processes take place (genome, transcriptome. proteome, etc.) will be simultaneously analyzed
yielding a much better view of the systems than that could be obtained separately.
Most of the ideas leading to our new methods and procedures have been motivated by our close and long standing relationship with
colleagues in severa! research institutions (The Flght against AIDS Foundation, Unidad de Patologla Vascular Cerebral del Hospital Clinic de Barcelona, Van dHebron lnstitut de Recerca, among others). 11 is of utrnost importance for us to enhance this relationship by providing solutions that taclde the right question and give a good solution. In line with this intention, a great ernphasis wil'I be given not only to
implementing !he methods developed but also in building user friendly interfaces to be accessible to a wide range of biomedical
anotaciones biológicas, análisis de supervivència, análisis integrativo de datos ómicos, asymptotic relative efficiency, bioequivalence trials, biological annotations, censura por intervalo, clinical trials, eficiencia relativa asintótica. ensayos de bioequivalencia, ensayos clínicos, estadistica multivariante, integrative analysis, interval censoring, modelos multiestado, multistate models, multivariate statistics, omics data, survival analysis
Adm. Estat
Plan Estatal de Investigación Científica y Técnica y de Innovación 2013-2016
Resoluton year
Funcding program
Programa Estatal de I+D+i Orientada a los Retos de la Sociedad
Funding call
Retos de Investigación: Proyectos de I+D+i
Grant institution
Gobierno De España. Ministerio De Economía Y Competitividad, Mineco


  • Gomez Melis, Guadalupe  (scientific coordinator)
  • Oller Sala, Josep Maria  (researcher)
  • Ocaña Rebull, Jordi  (researcher)
  • Julià de Ferran, Olga  (researcher)
  • Sánchez Pla, Àlex  (researcher)
  • Serrat Pie, Carles  (researcher)
  • Cobo Valeri, Erik  (researcher)
  • Vegas Lozano, Esteban  (researcher)
  • Langohr, Klaus  (researcher)
  • Perez Alvarez, Nuria  (researcher)
  • Kim, Kyungman  (researcher)
  • Hough, Guillermo  (researcher)
  • Dafni, Urania  (researcher)
  • Besalú Mayol, Mireia  (researcher)
  • Cortes Martinez, Jordi  (researcher)
  • Gómez Mateu, Moisés  (researcher)

Scientific and technological production

1 to 37 of 37 results