In silico characterisation of antigen receptor binding site structures
<p>The adaptive immune system defends the host against the invasion of foreign molecules known as the "antigens". These antigens are recognised by two main types of antigen receptors: T-cell receptors (TCRs) and antibodies. While both proteins share a globally similar β-sandwich arch...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
2020
|
| Subjects: |
| _version_ | 1826299496705818624 |
|---|---|
| author | Wong, WK |
| author2 | Deane, CM |
| author_facet | Deane, CM Wong, WK |
| author_sort | Wong, WK |
| collection | OXFORD |
| description | <p>The adaptive immune system defends the host against the invasion of foreign molecules known as the "antigens". These antigens are recognised by two main types of antigen receptors: T-cell receptors (TCRs) and antibodies. While both proteins share a globally similar β-sandwich architecture and are encoded by similar genetic mechanisms, TCRs are polyspecific and have medium affinity to their peptide antigens, while antibodies are highly specific and have high affinity to their targets. Their different behaviours are thought to be at least partially dictated by their binding site features. In this thesis, we aim to analyse their binding site structures and develop tools that can leverage the greater breadth of binding site diversity unveiled by repertoire sequencing. </p>
<p>In both types of proteins, the majority of the binding site is constituted by the complementarity-determining region (CDR) loops. In this thesis we first describe the development of a rapid sequence-based canonical form prediction tool (SCALOP) for antibody CDRs and for TCR CDRs. Based on this initial structural annotation, we then explored the structural differences between antibody and TCR CDRs and found that TCR CDRs tend to adopt multiple conformations, more often than their antibody counterparts. To capture the potential ensemble of binding site conformations, we built a TCR modelling tool, TCRBuilder. </p>
<p>Moving on from the structure of CDR alone, we then developed Ab-Ligity, a structure-based method that identifies sequence-dissimilar antibodies against the same epitope. This method incorporates the predicted antibody structure and the physicochemical properties of the binding site. Finally, as Ab-Ligity is dependent upon prediction of the paratope, we evaluated the leading paratope prediction method, Parapred. We attempt to highlight possible features that could portray paratopes with interpretable statistical models. The thesis concludes with the potential future directions of the work on binding site analysis for antibodies and TCRs.</p> |
| first_indexed | 2024-03-07T05:02:49Z |
| format | Thesis |
| id | oxford-uuid:d8e36add-7a2c-4e79-b174-5d64937e9440 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T05:02:49Z |
| publishDate | 2020 |
| record_format | dspace |
| spelling | oxford-uuid:d8e36add-7a2c-4e79-b174-5d64937e94402022-03-27T08:52:05ZIn silico characterisation of antigen receptor binding site structuresThesishttp://purl.org/coar/resource_type/c_db06uuid:d8e36add-7a2c-4e79-b174-5d64937e9440Structural bioinformaticsEnglishHyrax Deposit2020Wong, WKDeane, CM<p>The adaptive immune system defends the host against the invasion of foreign molecules known as the "antigens". These antigens are recognised by two main types of antigen receptors: T-cell receptors (TCRs) and antibodies. While both proteins share a globally similar β-sandwich architecture and are encoded by similar genetic mechanisms, TCRs are polyspecific and have medium affinity to their peptide antigens, while antibodies are highly specific and have high affinity to their targets. Their different behaviours are thought to be at least partially dictated by their binding site features. In this thesis, we aim to analyse their binding site structures and develop tools that can leverage the greater breadth of binding site diversity unveiled by repertoire sequencing. </p> <p>In both types of proteins, the majority of the binding site is constituted by the complementarity-determining region (CDR) loops. In this thesis we first describe the development of a rapid sequence-based canonical form prediction tool (SCALOP) for antibody CDRs and for TCR CDRs. Based on this initial structural annotation, we then explored the structural differences between antibody and TCR CDRs and found that TCR CDRs tend to adopt multiple conformations, more often than their antibody counterparts. To capture the potential ensemble of binding site conformations, we built a TCR modelling tool, TCRBuilder. </p> <p>Moving on from the structure of CDR alone, we then developed Ab-Ligity, a structure-based method that identifies sequence-dissimilar antibodies against the same epitope. This method incorporates the predicted antibody structure and the physicochemical properties of the binding site. Finally, as Ab-Ligity is dependent upon prediction of the paratope, we evaluated the leading paratope prediction method, Parapred. We attempt to highlight possible features that could portray paratopes with interpretable statistical models. The thesis concludes with the potential future directions of the work on binding site analysis for antibodies and TCRs.</p> |
| spellingShingle | Structural bioinformatics Wong, WK In silico characterisation of antigen receptor binding site structures |
| title | In silico characterisation of antigen receptor binding site structures |
| title_full | In silico characterisation of antigen receptor binding site structures |
| title_fullStr | In silico characterisation of antigen receptor binding site structures |
| title_full_unstemmed | In silico characterisation of antigen receptor binding site structures |
| title_short | In silico characterisation of antigen receptor binding site structures |
| title_sort | in silico characterisation of antigen receptor binding site structures |
| topic | Structural bioinformatics |
| work_keys_str_mv | AT wongwk insilicocharacterisationofantigenreceptorbindingsitestructures |