After her PhD, Thornton worked in molecular biophysics with David Chilton Phillips at the University of Oxford.[19][20] In 1978, she returned to the National Institute for Medical Research, and following that took up to a Fellowship at Birkbeck College, part of the University of London. In 1990 she was appointed Professor and Director of the Biomolecular Structure and Modeling Unit in the Department of Biochemistry and Molecular Biology at University College London and later also was appointed to the Bernal Chair in the Crystallography Department at Birkbeck College.[citation needed]
Thornton's work is highly interdisciplinary, interfacing with structural biology, bioinformatics, biological chemistry and chemoinformatics, amongst others. She was an early pioneer in structure validation for protein crystallography, developing the widely used ProCheck software.[23] Together with Christine Orengo, she introduced the CATH[24] classification of protein structure.[4][6][25][26][27][28][29][30][31] Her group developed a robust enzyme classification, comparison and annotation tool – the EC-BLAST[32] which calculates similarity between enzymes based on chemical reactions by capturing the bond change(s), reaction centre(s) or structural similarity between them.[33][32]
Janet Thornton is distinguished for her contribution to understanding protein three-dimensional structure: her perceptive comparative studies have led to the development of algorithms that are used to analyse and make predictions of supersecondary and tertiary structure. In the 1970s at Oxford (with M J Sternberg) she established clear and useful rules for the handedness of B-a-B units and demonstrated valid methods for prediction of the ordering of strands in B-sheets. At Birkbeck she developed this work to define families of conformations in B-hairpins and aB-links where the structures had previously been assumed at random. She has made the most comprehensive and useful analyses of tertiary interactions of protein sidechains, leading to an atlas that is valuable for protein and ligand design. The atlas is used widely in both academia and the pharmaceutical industry. At University College she has developed studies of sidechain conformation and stereochemistry into a procedure, PROCHECK, for evaluating the quality of experimentally defined protein structures: this is used widely to check protein structures. She has presented a method, known as threading, which gives strong evidence about tertiary structure for a protein sequence which is not obviously homologous to any other known structure.[40]
Dame Janet Thornton is Director of the European Bioinformatics Institute and is a world leader in bioinformatics. She has contributed significantly to medical science by increasing our fundamental understanding of the structure of proteins and how they contribute to disease and ageing. The tools and databases she has developed are used worldwide for basic research, in academia and also in pharmaceutical companies.
As Director of the EBI, she has been responsible for strategic developments related to the impact of the life sciences data on medical science. She is actively pursuing the challenge of how to join up biological and medical data in the UK and building tools which will facilitate the exploitation of these data for research and in the clinic.[41]
^Thornton, J. M.; Bayley, P. M. (1976). "Conformational energy calculations for dinucleotide molecules. A systematic study of dinucleotide conformation, with application to diadenosine pyrophosphate". Biopolymers. 15 (5): 955–975. doi:10.1002/bip.1976.360150511. PMID177120. S2CID30211950.
^Perkins, W. J.; Piper, E. A.; Thornton, J. (1976). "Computer techniques for conformational studies of biological molecules". Computers in Biology and Medicine. 6 (1): 23–31. doi:10.1016/0010-4825(76)90034-2. PMID1253578.
^Sibanda, B. L.; Blundell, T. L.; Thornton, J. M. (1989). "Conformation of beta-hairpins in protein structures. A systematic classification with applications to modelling by homology, electron density fitting and protein engineering". Journal of Molecular Biology. 206 (4): 759–77. doi:10.1016/0022-2836(89)90583-4. PMID2500530.