Editing Immune system (section)

=== Predicting immunogenicity ===
Some drugs can cause a neutralizing immune response, meaning that the immune system produces [[neutralizing antibodies]] that counteract the action of the drugs, particularly if the drugs are administered repeatedly, or in larger doses. This limits the effectiveness of drugs based on larger peptides and proteins (which are typically larger than 6000 [[Dalton (unit)|Da]]).<ref name="Baker2010">{{cite journal | vauthors = Baker MP, Reynolds HM, Lumicisi B, Bryson CJ | title = Immunogenicity of protein therapeutics: The key causes, consequences and challenges | journal = Self/Nonself | volume = 1 | issue = 4 | pages = 314–322 | date = October 2010 | pmid = 21487506 | pmc = 3062386 | doi = 10.4161/self.1.4.13904 }}</ref> In some cases, the drug itself is not immunogenic, but may be co-administered with an immunogenic compound, as is sometimes the case for [[paclitaxel|Taxol]]. Computational methods have been developed to predict the immunogenicity of peptides and proteins, which are particularly useful in designing therapeutic antibodies, assessing likely virulence of mutations in viral coat particles, and validation of proposed peptide-based drug treatments. Early techniques relied mainly on the observation that [[hydrophile|hydrophilic]] [[amino acid]]s are overrepresented in [[epitope]] regions than [[hydrophobe|hydrophobic]] amino acids;<ref name="Welling">{{cite journal | vauthors = Welling GW, Weijer WJ, van der Zee R, Welling-Wester S | title = Prediction of sequential antigenic regions in proteins | journal = FEBS Letters | volume = 188 | issue = 2 | pages = 215–18 | date = Sep 1985 | pmid = 2411595 | doi = 10.1016/0014-5793(85)80374-4 | doi-access = free | bibcode = 1985FEBSL.188..215W }}</ref> however, more recent developments rely on [[machine learning]] techniques using databases of existing known epitopes, usually on well-studied virus proteins, as a [[training set]].<ref name="Sollner">{{cite journal | vauthors = Söllner J, Mayer B | title = Machine learning approaches for prediction of linear B-cell epitopes on proteins | journal = Journal of Molecular Recognition | volume = 19 | issue = 3 | pages = 200–08 | year = 2006 | pmid = 16598694 | doi = 10.1002/jmr.771 | s2cid = 18197810 }}</ref> A publicly accessible database has been established for the cataloguing of epitopes from pathogens known to be recognizable by B cells.<ref name="Saha">{{cite journal | vauthors = Saha S, Bhasin M, Raghava GP | title = Bcipep: a database of B-cell epitopes | journal = BMC Genomics | volume = 6 | pages = 79 | year = 2005 | pmid = 15921533 | pmc = 1173103 | doi = 10.1186/1471-2164-6-79 | doi-access = free }}</ref> The emerging field of [[bioinformatics]]-based studies of immunogenicity is referred to as ''[[immunoinformatics]]''.<ref name="Flower">{{cite journal | vauthors = Flower DR, Doytchinova IA | title = Immunoinformatics and the prediction of immunogenicity | journal = Applied Bioinformatics | volume = 1 | issue = 4 | pages = 167–76 | year = 2002 | pmid = 15130835 }}</ref> [[Immunoproteomics]] is the study of large sets of proteins ([[proteomics]]) involved in the immune response.<ref name="pmid31356379">{{cite journal |vauthors=Kanduc D |title=From hepatitis C virus immunoproteomics to rheumatology via cross-reactivity in one table |journal=Current Opinion in Rheumatology |volume=31 |issue=5 |pages=488–492 |date=September 2019 |pmid=31356379 |doi=10.1097/BOR.0000000000000606|s2cid=198982175 }}</ref>