Figure 3. Network representation of the associations between allergens’ entry way and symptomatology. The node labelled as “–” represents the allergens without annotated entry way, that correspond mainly to allergens of nematode origin.
There are many relationships with good p-values and they make sense from a biological/clinical point of view. Moreover, the symptomatology associated to a given exposure route tends to be specific to it, with only a few symptoms shared between routes (Figure 3) . As mentioned above and now quantified, allergens with the “Airway” exposure route are clearly associated to symptoms and pathological terms intuitively related to the respiratory system, such as “asthma”, “lung disease” or “rhinitis”. Similarly, “Dermal-skin” is associated with latex-related allergy, eczema, dermatitis, etc. We also find relationships between “Dermal-skin” and spina bifida related pathological terms, due to the relationships between latex allergens and this pathology. This relationship has been repeatedly reported in the literature: patients with spina bifida have a higher incidence of latex allergy, probably due to the exposure to latex in surgical procedures and implants[23]. So, this relationship is correct but in the opposite direction to the others we try to detect: it is not latex allergy what causes spina bifida but the other way around. The “Food” entry way is related to the trivial “Food allergy” (HP:0500093), “food-induced anaphylaxis”, nut-related hypersensitivity, celiac disease, body weight problems, etc. Finally, the “Injection” entry way is related to “poisoning”, “chemically-induced disorders”, “hemolysis”, “insect allergy”, etc (Figure 3 ).