Advertisement

Cyrus Chothia obituary

Cyrus Chothia obituary. Biochemist whose work was at the cutting edge of the understanding of protein structures, their function and evolution

Taxonomy – the classification of objects according to their relationships to one another – conjures up images of 19th-century amateur naturalists measuring fossils or counting the stamens of flowering plants.

The biochemist Cyrus Chothia, who has died aged 77, took a taxonomic approach to research at the cutting edge of molecular biology, organising the bewildering variety of protein structures revealed by techniques such as x-ray crystallography and genome sequencing into coherent family trees.

Employing computational analysis, he laid much of the groundwork for the field of bioinformatics, which now underpins the study of biodiversity, the global search for links between genes and disease, and the rational development of new drugs.

Beginning in the 1950s, and accelerating from the 70s as the power and speed of computers improved, techniques such as x-ray crystallography made it possible to visualise the three-dimensional structures of protein molecules and so begin to understand how they interact in living systems. But as long as only a few dozen structures were known, scientists were bewildered by the apparently arbitrary way linear protein chains folded up into their three-dimensional shapes.

From the moment he became established as an independent scientist, Chothia, with his close colleague Arthur Lesk at the Laboratory of Molecular Biology (LMB) in Cambridge, took a computational approach to analysing the relationships between the sequences and structures of proteins. A particularly fruitful area was their work on antibodies, the large proteins that form part of the armoury of the immune system by recognising invading bacteria and viruses.

Chothia and Lesk showed that the apparently limitless variety of antibodies is created in the body using only a limited repertoire of structural conformations. They were able to predict and classify antibody structures from their linear sequences, a finding that has been of critical importance in developing antibody therapies.

From the early 90s the success of this approach led Chothia to contemplate the wider protein universe. Working with colleagues at the LMB, he set up the Structural Classification of Proteins (SCOP) online database. Beginning by simply comparing protein structures visually, and moving on to computer automation, they arrived at a hierarchy of subfamilies, families and superfamilies defined by similarities of structure and function.

With this tool to hand, scientists who find new proteins can assign them to families and immediately begin to grasp their possible functions, or infer from the structure of a protein in a mouse, for example, what the function of a similar protein in a bacterium or a human might be.

Chothia began this work just as projects to sequence the genomes of species from yeast to human were getting under way. In 1992 he speculated, on the basis of similarities among the relatively small number of proteins whose genes had then been sequenced, that there might be no more than 1,000 protein families in total. The current version of SCOP, which includes over half a million proteins, classifies them into around 2,500 evolutionarily related families – relatively speaking, not that far off from Chothia’s original estimate.

The data confirm and quantify an observation made on some of the first protein structures to be solved: that molecular shapes can be much more enduring than protein sequences over the course of evolution, and provide a sounder basis for understanding evolutionary relationships. They also show that as living organisms evolve to be more complex, they produce increasingly complex protein structures, formed from combinations of simpler proteins.

Chothia’s commitment to bioinformatics (he preferred the phrase “theoretical and computational biology”), and to creating tools that can be shared by the whole research community free of charge, has enabled what began as a largely descriptive science to acquire predictive power, possible protein structures being almost instantly predicted from their genetic sequences. For this work he was awarded a share of the $1m international Dan David prize.

Chothia was born in Windsor, Berkshire, the son of Homi Chothia, a Parsi from Bombay (Mumbai), India, who emigrated to Britain in 1932 to learn to fly and who reached the rank of flight captain during the second world war delivering planes for the Air Transport Auxiliary. His mother, Betty (nee Foskett), was a beautician. The elder of two sons, Cyrus attended Alleyn’s school in Dulwich, south London, and went to the University of Durham to study chemistry, graduating in 1965.

Having learned the technique of x-ray crystallography during a master’s at Birkbeck College, London, he did his PhD at University College London. His supervisor, Peter Pauling, worked on the structure of neurotransmitters and psychoactive drugs such as LSD. Pauling, the son of the American double Nobel prizewinner Linus Pauling, was also pioneering computer-based approaches to visualising molecular structures.

Chothia’s PhD project included solving the structure of the neurotransmitter acetylcholine and led to a three-year fellowship at the LMB, the mecca of structural studies, where he began work on proteins. For the next three years he visited labs at Yale, the Weizmann Institute in Israel and the Pasteur Institute in Paris, and it was during this period that he first began to work out the principles that would enable him to classify proteins by their structural composition and relationships. On his return he held joint posts at UCL and the LMB until 1990, when he became a group leader at the LMB.

A former student remembers that he regarded the LMB as “the centre of the universe”, and never sought a post elsewhere. He avoided commitments that would take him away from his lab, rarely attending scientific conferences.

Like most LMB staff he ran a small group, with no more than one research student per year. He was scrupulously fair in his dealings with his colleagues, giving students full credit for their published papers, and was notably supportive of female researchers at a time when it was far from the norm. He continued working into his early 70s, when ill health forced him to step back.

Chothia met his wife, Jean (nee Sandham), through shared political activities while they were both undergraduates at Durham, and they married in 1967. He was interested in cinema, architecture, classic photography, travel writing and fine printing and was a passionate book collector in all these areas.

Jean, who specialises in drama and performance as reader emerita in English at the University of Cambridge, survives him, as do their children, Lucy and Tom, and three grandchildren.

• Cyrus Homi Chothia, biochemist, born 19 February 1942; died 26 November 2019