IMforFUTURE

Training the next generation of omics researchers Dr Jeanine Houwing-Duistermaat of the IMforFUTURE (Innovative Training in Methods for Future Data) project is training and guiding the next generation of omics researchers to establish the most effective methods to measure, integrate, and analyse datasets.

IMforFUTURE is focused

on omics research, which involves several disciplines.

Omics research is seen as having enormous potential for healthcare. The different omics disciplines, when combined, can paint a holistic portrait of an individual’s precise health and healthcare needs.

Omics datasets enable the gathering of novel insights into stages in biological processes. Studies include genome-wide DNA markers reflecting the genetic code, transcriptomics that quantifies expression of genes, proteomics measuring the abundance of proteins, and glycomics that studies sugar molecules surrounding and modifying proteins in your body. IMforFUTURE is directing a lot of attention to scrutinising glycomics. Glycoscience is expected to be a key to realising personalised medicine goals.

In the future, it may be the case we have highly personalised healthcare, that is predictive, not reactive, which is bespoke to an individual’s genetic makeup, metabolism and the myriad of processes that occur within their body at the most fundamental levels of their biology. It will be understanding the complete nature of a person’s biological and chemical

processes and how they will react to medicines and interventions. In turn this will lead to the most effective treatments and show us how we could reverse ageing processes. In medicine, we currently often have a one size fits all mentality, but always knowing how the body will react could lead to more customised and personalised treatments.

often too time-consuming. The omics markers are often analysed one by one, ignoring the joint involvement of several markers and the complexity of measurement techniques.

Omics data, while representing the same processes in the body, are very different in biological and chemical properties. Models that integrate the omics data and also

Dr Hae Won Uh and I became aware of the necessity of an interdisciplinary training network, because of several missteps analysing omics data due to a lack of understanding of each other’s disciplines. Cross-disciplinary skills should prevent this type of mistake. We set up the IMforFUTURE training programme in chemistry, epidemiology and statistics.

Integration of data and knowledge Before this healthcare vision becomes a reality, it is a research prerogative to find high throughput methods to measure the omics markers accurately and statistical methods for identifying the relevant molecular profiles from these data. Experimental methods are

address these differences, perform better than methods ignoring these differences.

Measuring, analysing and interpreting data are linked with each other in one workflow.

For a data scientist it is crucial to know: What type of data is in the database? How were the samples organised by the chemist when they perform the measurements? How

was the quality of a single measurement determined? On the other hand, the chemist needs to know that their handling of samples may affect down-stream analysis of the data.

For example, if a lab technician decides to measure a few samples twice, this information is relevant for the data scientist in choosing the right model. Ignoring this information may lead to false findings.

“Dr Hae Won Uh and I became aware of the necessity of an interdisciplinary training network, because of several missteps analysing omics data due to a lack of understanding of each other’s disciplines. Cross-disciplinary skills should prevent this type of mistake. We set up the IMforFUTURE training programme in chemistry, epidemiology and statistics.”

Multidisciplinary training of IMforFUTURE The programme comprises three components for fellows to develop themselves as multidisciplinary researchers: course modules, secondments and research at their host institutes.

The course modules were organised in four sessions, and each time there were issues arising from multiple disciplines, they were addressed. In practice, this meant that the fellows followed lectures which dealt with materials they had already covered in their bachelor programmes, while other lectures contained completely new materials for them.

The training was followed by an assignment that tested knowledge and skills from all the involved disciplines. Fellows were allowed to

OmicsPLS workflow

help each other but had to formulate their own answers. This format worked well. By working together, fellows formed a group and at a later stage, when performing their own research, they were able to get input from other disciplines by contacting the other fellows.

The fellows had to do at least one secondment in a complementary institute.

The data-scientists spent one month in a laboratory to learn how the data are measured, which they analysed later. The chemists went with their data to datascience institutes to learn about the ins and outs of analysis methods. By doing so, the data scientists got acquainted with all kind of experimental errors, and the chemists learned how experimental error could be accounted for in the analysis.

Finally, fellows were expected to perform

research in an interdisciplinary environment.

A success story is Zhujie Gu, a statistician, who visited Genos in Zagreb and the University of Bologna, to study glycomics and to learn about healthy ageing. He has now ongoing collaborations with five different omic groups, resulting in two publications, a software package OmicsPLS and another three manuscripts in progress.

Age is not just a number “I am a statistician, and the methods which I develop are useful for many domains. In IMforFUTURE we are interested in ageing.

One of our partners developed GlycanAge. You can estimate how old someone is in biological age, rather than chronological age. This can be useful if you are elderly, and a decision needs to be made about whether surgery is a good idea or not. You don’t want

Cartoon artist: Pollie Hogenboom. The cartoon was financed by the FP7 project MIMOmics no 305280.

16

EU Research

www.euresearcher.com

17

IMforFUTURE Innovative training in methods for future data

Project Objectives

IMforFUTURE (Innovative Training in Methods for Future Data) is an innovative multidisciplinary and intersectoral research training programme which addresses current shortcomings in omics research. We aim to open the new research horizon in integration of genetics, glycomics, and epigenomics datasets into systems biology by developing innovative methods for high throughput omics and for integrative analysis of omics data. We focus on ageing, which is the biggest single risk factor for many diseases.

By application of our novel methods to emerging datasets representing inflammation and immunology, IMforFUTURE will contribute to understanding of the underlying biological processes involved in diseases and ageing.

Project Funding

This project has received funding from the European Union’s Horizon 2020 research and innovation programme, under H2020MSCA-ITN grant agreement number 721815.

Project Partners

• https://imforfuture.eu/partners/

Contact Details

Project coordinator, Professor Jeanine Houwing-Duistermaat School of Mathematics University of Leeds Leeds LS2 9JT T: +Xxxxxxxxxxxxxx E: IMforFUTURE@leeds.ac.uk W: www.imforfuture.eu Bouhaddani, S., Uh, HW., Jongbloed, G. et al. Integrating omics datasets with the OmicsPLS package. BMC Bioinformatics 19, 371 (2018). https://doi.org/10.1186/s12859-018-2371-3.

Krištić J, Vučković F, Menni C, et al. Glycans are a novel biomarker of chronological and biological ages. J Gerontol A Biol Sci Med Sci. 2014;69(7):779789. doi:10.1093/gerona/glt190

Professor Jeanine Houwing-Duistermaat

Jeanine Houwing-Duistermaat is professor of data analytics and statistics at the University of Leeds. She is passionate about data. Her research interests include statistical bioinformatics, in particular modelling of multiple omics datasets simultaneously. She works with chemists, biologists, epidemiologists and clinicians.

18

to operate on them if there is a chance of them dying or if quality of life is at risk, as that would make no sense.

“The concept is that if you give a drop of blood, your biological age can be determined and if it’s much higher than your chronological age then it can indicate you would benefit to change your lifestyle. At the moment, this is based on what is available in the GP records, but predicting biological age involves mixed datasets.”

The multi-dimensional human body Combining the measuring of data and analytics of that data in omics research is showing possibilities and potentials but there is plenty of work to do.

silo when the body is so complex makes it easy to understand how limited understanding leads to inefficient methodologies. Enabling researchers to swap insights, knowledge and understanding, as Jeanine puts it, “leads to them generating new questions.” She explains: “That is what you want at the end of the project, it’s good to come to a place when new questions are generated. For example, how do we include existing related data and information in our models? Can we develop a model to support decision making in the laboratory with regards to the most promising experiments? Can we reveal information on the structure and dynamics of the molecules from current datasets?”

IMforFUTURE is in essence, a multidisciplinary training programme

I think the fact that they are multi-disciplinary and that they know how to analyse and interpret the data and they know what they don’t know but how to ask the right questions about data to the right expert, that will make them attractive for employers.

“There is a lot of information coming out of these machines, more than the chemists often realise. I think the people from the different disciplines should learn and know much more from each other to extract all the information. The fellows have seen that we can be more efficient by having some knowledge of the other fields. You need to spend time to understand each other, of course. This goes much further than statistical consulting where we help each other with a goal to have novel results in one discipline, instead of in all the involved disciplines, to really make a difference. One of the problems with precision medicine is, we need more ideas to understand and address the complexity of what we measure and how we analyse that. We should realise that we make something one-dimensional while it is four-dimensional. We often lack information on structure and dynamics. How to get the whole picture?”

Bringing fellows from different omics disciplines together was a key aim for IMforFUTURE. Combining knowledge, a better, fuller picture emerges. Operating in a

for omics researchers to ensure they are competent in measurement techniques of underlying mechanisms, can design studies and are able to analyse data. By networking the new generation of scientists in these fields, for developing cross sector and cross discipline understanding, it will make them highly employable.

“I think the fact that they are multidisciplinary and that they know how to analyse and interpret the data and they know what they don’t know but how to ask the right questions about data to the right expert, that will make them attractive for employers.”

Typical profile of glycans of the IgG glycome.

EU Research