IBM Research-Zurich

Microfluidic probe for personalized healthcare

Share this post:

The sequencing of the entire human genome back in the early 2000s is considered one of the most important scientific breakthrough enabling data-driven medicine.

It flooded researchers in biology with a tremendous amount of Big Data that needed to be processed and tested. It was around then that DNA microarrays, small glass chips where thousands of DNA fragments are deposited, became popular because they enable high-throughput and multiplexed testing for gene expression and mutations, for example in cancer.


From left to right, Julien Cors, Govind Kaigala and David Taylor

Interestingly, DNA and protein microarrays were one of the convergence points of microtechnology and biology that resulted in the fabrication of high-density arrays of biomolecules.

While protein microarrays have significant value for drug discovery, and molecular profiling, the same success in the use of DNA arrays was not transferable to protein arrays due to the limited quality of protein arrays.

This inspired a team of IBM scientists to develop a new biopatterning method for efficient, accurate and high-quality patterning of proteins on surfaces by addressing fundamental bottlenecks inherent to the fabrication of protein microarrays. Such high-quality microarrays will benefit quantitative biological assays for personalized diagnostics and screening applications.

A new paper appearing this month on the cover of Analytical Chemistry details their latest research breakthrough. I spoke with two of the authors Govind Kaigala and Julien Cors.


One of the benefits of your technique is depositing reagents on surfaces in a wet environment. Why is this important and how does it compare to current approaches?

Govind Kaigala (GK): Current techniques for protein patterning use inkjet and pin spotters, and they do not produce homogeneous protein patterns — largely because they operate in a dry environment.

So we asked ourselves, biology is wet, so how can we accurately test samples when we are changing it’s make up by drying it out?

In contrast, our method relies on a microfluidic probe (MFP) that confines nanoliter volumes of reagents in a wet environment. By fully immersing the substrate in a liquid, the “coffee stain effect” (Marangoni effect), is minimized with the reduction of evaporation. In addition, we use continuous flows on the surface, which ensure improved reaction kinetics compared to deposition relying on diffusion in existing techniques. Also via a technique of re-circulating liquid we improve reagent utilisation, which is important because its very expensive.

In our article we studied the transport of biomolecules in flows and their surface interactions and to address this we designed specific MFP heads to create homogeneous protein patterns with volume consumption comparable to inkjet spotters.

The paper also describes the convection-enhanced transport and recirculation of sub-microliter volumes using analytical models.

A major factor of your research is working at the micrometer level. Can you explain what the trick is when working with such small sample sizes?

Julien Cors (JC): IBM is considered the birthplace for scanning probe techniques with the original scanning probes developed for atomic imaging. Our scanning probe isn’t imaging atoms, the MFP technology that we are developing is for biological applications, in this case protein patterning. But, the technique is similar in that the probe head never comes into contact with the sample and is used in liquid environments and with a resolution of a few micrometers.

By continuously injecting and aspirating liquids, nanoliter volumes of liquids are confined at the tip of the head. To put this into perspective, we are working with volumes of liquids thousands of times smaller than a tear drop.

At this micrometer-length scale both reaction kinetics and transport of biomolecules onto the surface are highly favourable  By exposing one or multiple such confined liquid using the probe on a surface diverse patterns of biomolecules can be created.

Can you talk about some specific applications for the innovation?

GK: Protein-based assays (immunoassays) are very broadly applicable, and our method can be used to create high-quality substrates to implement such tests.

Current protein microarrays provide qualitative information i.e. presence or absence of a specific biomarker. Because of the inhomogeneity of the deposited patterns, extracting quantitative information remains challenging. Quantitative data have great importance both in diagnostics and biomarker discovery.


For example, in the Analytical Chemistry paper we demonstrate an assay for the detection immunoglobulins (IgGs). IgGs are used as a diagnostics marker for several autoimmune diseases and as a measure of the immune response.

This research is partially funded by your European Research Council (ERC) grant. What do you hope to achieve by the end of the grant?

GK: Within the ERC-BioProbe project, we are developing new concepts, tools and methods for improved molecular profiling of tissues for tumor diagnosis for personalized medicine.

Techniques for liquid recirculation and analytical models for improved protein microarrays was in fact an off-shoot of this work – while in BioProbe, we will focus on the main theme, we keep an eye for such projects and pursue them.

What’s next for your research and when can we expect it to be used outside of IBM by partners?

GK: We are very excited about the progress and evolution of the microfluidic probe, particularly as it is now starting to be applied in various areas including pathology and personalized medicine.

Convection-Enhanced Biopatterning with Recirculation of Hydrodynamically Confined Nanoliter Volumes of Reagents, Julien AutebertJulien F. CorsDavid P. Taylor, and Govind V. Kaigala, Anal. Chem., 2016, 88 (6), pp 3235–3242, DOI: 10.1021/acs.analchem.5b04649

More IBM Research-Zurich stories

Linear-Complexity Earth Mover’s Distance Approximations for Efficient Similarity Search

The Earth Mover’s Distance is a highly discriminative metric for measuring distance between probability distributions that has been applied successfully in various fields.

Continue reading

Imaging molecules in different charge states

IBM researchers, along with collaborators at the Universidade de Santiago de Compostela and ExxonMobil, reported in the peer-review journal Science that they have been able to resolve with unprecedented resolution the structural changes of individual molecules upon charging.

Continue reading

Hypertaste: An AI-assisted e-tongue for fast and portable fingerprinting of complex liquids

For the rapid and mobile fingerprinting of beverages and other liquids less fit for ingestion, our team at IBM Research is currently developing Hypertaste, an electronic, AI-assisted tongue that draws inspiration from the way humans taste things.

Continue reading