Finding antibodies that are close sequence relatives to previously characterized antibodies with known binding partners is frequently performed to identify antigen-specific sequences.In order to find such shared antibodies, one needs to curate a set of previously characterized antibody sequences, typically from scientific publications and patents. Sequence reporting is heterogeneous and it is cumbersome to find publicly available sets of antibody sequences.
We gather data on naturally sourced and therapeutic antibodies to develop statistical models of both, to act as a reference to a potential antibody therapeutic candidate.
Antibodies are a class of proteins that evolved to identify noxious molecules in many vertebrates. Binding sites of these molecules are extremely diverse and one of the key questions in antibody engineering is finding a particular sequence, out of many possible ones, targeting a specific molecule's surface.
This is often performed by comparing a novel antibody to a previously characterized one with a known binding mode, on assumption that they would share specificity. In order to find such shared antibodies, one needs to curate a set of previously characterized antibody sequences.Sequence reporting is heterogeneous and it is cumbersome to find publicly available sets of antibody sequences. One of the most common outlets for deposition of sequence information is NCBI GenBank. To address the issue we curated databases of antibody sequences from the leading resources such as GenBank, called AbGenBank available at naturalantibody.com/abgenbankWe hope that our databases would facilitate myriad antibody analytics and discovery protocols that need to reference their molecules against those previously characterized.
Identifying previously characterized antibodies is an important step in many antibody analysis pipelines. Finding close-sequence matches from either patents or scientific publications can provide an indication whether a query antibodies (e.g. from Next Generation Sequencing) were previously identified. If the target of the previously characterized antibody is known, it could provide indication of shared specificity to the query antibody. Providing easy-to access search facilities to antibodies from patented sequences (PAD) and those from GenBank (AbGenBank) should benefit antibody analytics pipelines.Both PAD and AbGenBank are available for academic non-commercial use.
We publish our research results with many collaborators across academia and industry.
Our collaborators include:
Oxford UniversityCambridge UniversityUtrecht UniversityDanish Technical University
AstraZenecaBoehringer IngelheimUCB Pharma