Maarten van Schaik

Statistician developing mixed effect models for measurement error and overdispersion in correlated samples

Background

Hello, my name is Maarten van Schaik and I am a PhD student at University of Leeds at the School of Mathematics. I received my BSc in Criminology and MSc in Statistical Science for Life Sciences at Leiden University in the Netherlands. During my Master’s I had the opportunity to learn about and apply a diverse set of statistical methods, from including methods from the fields of traditional statistics, to machine learning and artificial neural networks. During my Master’s I also taught classes in statistics to undergraduates at the Faculty of Social and Behavioural Sciences of Leiden University. In January 2018 I moved to the United Kingdom to start my work as Early Stage Researcher within the IMforFUTURE Marie Skłodowska-Curie Innovative Training Network, under the supervision of Prof. Jeanine Houwing-Duistermaat and Prof. Arief Gusnanto.  

The human microbiome

the human microbiomeUsing modern 16S rRNA sequencing techniques, the human gut microbiota has become a new field of research. The scientific community is becoming increasingly aware of the importance of the gut microbiome for human health and longevity. It is believed that among other omics field such as genetics, proteomics and glycomics, a “healthy gut” microbiome plays a key role in healthy ageing. 

Statistics

As an ESR in IMforFUTURE, I am working on the development of novel statistical methodology to analyse the multivariate count data that are encountered in high throughput omics research such as the human gut microbiome. Modeling these counts it not an easy task. After mapping the bacterial RNA to a reference library and binning them into Operational Taxonomic Units, we are left with samples which are comprised of a very high number of different species per sample (high-dimensional data) and these data are highly variable by person, environment and time. 

My project aims to develop new methods while working with data from differential human gut microbiome assemblages during soil transmitted helminth infections.

Secondments

Maarten on secondment at glyXera with Frania explaining the lab work

During my secondment at glyXera GmbH I had the opportunity to learn from colleague-ESR Frania Zúñiga about her work with low-abundance glycoproteins. It was very interesting to see the work of the wet lab from a statisticians’ point of view.

In September-November 2019 I have worked at the University of Bolgona in Italy, doing joint research with colleague-ESRs Iva Budimir and Zhujie Gu on statistical modeling of omics data.

 

Collaborative work

Relative intensity and coefficient of variation broken down by N-glycopeptide.
Relative intensity and coefficient of variation broken down by N-glycopeptide. The results are based on the 4 technical replicates of blood plasma. The results show that many structures are detected, but only a few N-glycopeptide account for most of the relative signal (as measured by the integrated AUC of the XIC). Furthermore, more highly abundant N-glycopeptides tend to have lower CV (suggesting more confidence in presence in the replicates)

After the secondment at glyXera, I continued working with Frania on a collaborative project, culminating in a technical report on the variability in N-glycopeptide workflow. The goal of this report is to determine the technical variation of a workflow performed to reveal the N-glycosylation of blood plasma proteins. The report can be read on my GitHub pages website, and the the files for the statistical analysis are hosted on a GitHub repository.

More information

LinkedIn