We had several ideas about how the sensor could work, and eventually, one rose up strongly. The first step is having an idea. It may seem simple, but it can take long, as it goes through think about the idea, understand it and rethink it as many times as it is required. Maybe this reflection seems intuitive, but constitutes the fundamental part of all developments. Using mathematical models and computer simulations we have been able to understand the dynamics underlying our sensor. Furthermore, using this model we can tune a huge variety of parameters, test many eventualities and predict the tendency of the sensor.

One of the strengths of our approach to detect patient-specific reactive gluten peptides in food is that we can produce the sensor using only a DNA sample from the saliva of the patient, which is not invasive. Patient’s HLA-DQ (the protein responsible for recognizing and triggering the immune response) is then expressed using the PCR protocol designed in the WetLab. For this purpose, we use specific primers to amplify the patient’s target genes and thus, it is necessary to evaluate if they will be useful for all of the different celiac-associated genotypes.

Another aspect that we wanted to evaluate is how the HLA-DQ protein is able to effectively recognize and bind gluten-derived peptides. Specifically, which are the residues that are mediating this recognition and if they are conserved among celiac haplotypes DQ2 and DQ8.

To do so, we focused the conservation study to the residues that have been previously suggested to be responsible for the peptide recognition. A multiple alignment of protein amino acid sequences was performed using all the annotated HLA-DQ proteins so far.