To better support our staff in complying with the UK and Scottish Governments'
social distancing policies during the current COVID-19 pandemic, BioSS has
switched to a home-based, remote working model for our activities.
We have made strenuous efforts to ensure that our staff will be efficient and
effective while working from home, and our intention is to continue to operate
as best we can during the outbreak. Research and consultancy work is ongoing,
and training courses are now taking place online.
All BioSS staff are contactable by electronic means, and we are happy to arrange virtual meetings with new and existing collaborators.
BioSS is a founder member of the Scottish COVID-19 Response
Consortium which is developing open source
epidemiological models of COVID-19 spread designed to inform policy. BioSS
staff are contributing through the management team and via development of
detailed large scale simulation modelling, statistical inference of model
parameters from data and information processing pipelines.
Javier Palarea-Albaladejo
Principal Statistical Scientist
Biomathematics and Statistics Scotland
JCMB, The King's Buildings,
Peter Guthrie Tait Road,
EDINBURGH, EH9 3FD, Scotland, UK.
Development and application of statistical modelling and data analysis methods in interdisciplinary scientific research; methodological research; quantitative services for externally-funded projects; and training for scientists.
I am based at the BioSS Headquarters in Edinburgh, but also spend some days providing on-site support at the Moredun Research Institute. My areas of specialisation include:
Multivariate data analysis
Compositional data analysis
High-throughput data analysis for chemometrics and bioinformatics
Missing and censored data
Statistical computing, particularly R programming and development
Some current interdisciplinary projects
Nematode vaccine development: I have been working with scientists at Moredun Research Institute on the development of effective candidate vaccines against nematodes like
Teladorsagia Circumcincta This work, initiated under the SRP, is continuing within the PARAGONE project, an international partnership
funded by EU's Horizon 2020.
Development of prototype on-hen feeding device to optimise poultry red mite vaccine efficacy studies:
the key objective of this project is the assessment of novel methods of controlling poultry red mite in hens without performing large scale,
prolonged field vaccine trials (funded by the National Centre of the Replacement, Refinement & Reduction of Animals in Research, NC3R).
Host factors in determining resistance to cryptosporidiosis in cattle: this project uses an integrative approach to fully characterise the host-parasite interactions that influence the development of disease and
resistance to the Cryptosporidium parasite, which is crucial for the development of effective vaccine strategies and to identify relevant biomarkers to help control this disease in cattle (funded by BBSRC).
Predictive modelling of cattle methane emissions: this project combines methodological and interdisciplinary work with the Scotland's Rural College to explore the potential for using rumen
volatile fatty acids (VFA) and, more widely, high-throughput metabolomic profiles as proxies for cattle methane emissions.
Methodological research
My main area of interest is compositional data analysis and its applications. Compositional data essentially refer to multivariate data conveying relative information. This is commonly the case of data which are expressed, or can be meaningfully expressed, in units such as percentages, parts per million, mg/m3, minutes/day, or similar. Some examples include:
Chemical and nutritional compositions
Land use, budget, portfolio distributions
Behaviour, choice, activity, time use data
Abundance of species or organisms in e.g. environmental, microbiome and metagenomics studies
Relative information and the inherent interplay between parts of a composition are not generally considered by ordinary statistical methods. Hence, these can lead to both technical issues and partial or even completely misleading understanding of the processes governing the scientific question under study. I work on novel compositional models and methods as applied to diverse scientific areas like animal behaviour,
physical activity and human health, livestock methane production, food nutritional labelling and water catchment and environmental research.
Some related links:
CoDaWork 2019, The 8th international Workshop on Compositional Data analysis (3-8 June)
Another area of interest is the development of robust MALDI mass spectrometry data pre-processing algorithms and predictive models aimed at the accurate identification
and differentiation of bacterial strains like Escherichia coli and Campylobacter from complex proteomic spectral profiles.
This work was initiated with funding from the Scottish Government's Strategic Partnership for Animal Science Excellence (SPASE, 2011-16) and it is currently continued under the Scottish Government's Strategic Research Programme.
Software packages I have developed in relation to these areas include zCompositions and MALDIrppa.
Association between the VFA composition [propionate, Acetate x butyrate, others] and individual animal methane yield (CH4; grams per kg of dry matter intake in natural log scale) accounting for diet type (concentrate, mixed, forage).
The ratio of the combination of acetate and butyrate to propionate, which is analogous to the so-called glucogenic ratio, was found to be a key driver of the relationship.
