What do genes, medieval manuscripts, languages, Persian carpets and folktales have in common? - Fellows' seminar by Christopher Howe

“Evolution depends on DNA being copied accurately – but not completely accurately. Sometimes changes (mutations) occur, which provide the variation necessary for evolution. Biologists use the accumulated changes in DNA sequences to infer evolutionary history. However, DNA is not the only thing that undergoes copying with the incorporation of changes. The same is true for texts, copied by scribes before the development of printing. Recognising this, we have collaborated for some time with textual scholars to use computer programs from evolutionary biology to uncover the copying history of sets of texts. This has now expanded to look at other things like languages, Persian carpets and even folktales,” said Christopher Howe of the Department of Biochemistry and Corpus Christi College at the University of Cambridge. Howe took fellows on an entertaining journey incorporating some of the fascinating history of the Corpus Christi College which was founded in 1352 and contains the Parker Library – the finest  private collection of early English and Anglo-Saxon manuscripts including the Augustine Gospels from 597 AD which were used at King Charles’s coronation; to his work on photosynthesis and its potential to generate small amounts of green power; and, to the possibility of constructing phylogenetic or evolutionary trees for many different types of things.

A brief history of slime

He explained that the main work of his laboratory is on photosynthesis – how plants and algae turn light energy into chemical energy powering almost all life on the planet, and how it has evolved over billions of years. The first evidence of life – chemical signals in rocks and microfossils - date to 3.5 billion years ago. The first evidence of oxygen-producing photosynthesis is in stromatolites (layered sedimentary rock formations resulting from algae trapping sediments over time) dating to 3 to 2.7 billion years ago; while the origin of chloroplasts (the sub-cellular compartments responsible for photosynthesis in plants) dates to 1.2 billion years ago. Howe explained that photosynthesis is the process that plants undertake to create sugar and oxygen from carbon dioxide, water and sunlight. This takes place to feed the plant and ensure its survival, and in the process electrons are generated in the thylakoid membranes; some of these electrons can be converted to small amounts of ‘green’ electric energy. “This electric power can be used to power microprocessors and run small devices which is useful in off-grid locations. We have used electricity from plants to power a digital camera – taking photos of the plant – essentially a plant selfie.” “Overall, we study the evolution of photosynthetic micro-organisms,” said Howe. “And these studies provide some unexpected observations, including that one of the world’s most important pathogens has an algal ancestry.” Howe explained that the parasite that causes malaria - Plasmodium falciparum is closely related to a group of algae – dinoflagellates, which provide nutrition to coral. “This means an ancestor of Plasmodium could do photosynthesis but the parasite doesn’t anymore,” said Howe. “But it still may have related biochemical functions which might prove to be good targets for drug development.” “The ability to do photosynthesis is spread through the tree of life,” he continued. “And the ability to do photosynthesis has transformed organisms throughout the tree of life. Even Darwin pointed to tree-like evolution – with an evolutionary tree the only figure in Origin of the Species.” So, how do we construct such evolutionary trees? Howe explained that organisms that are closely related (or share recent ancestry) are more similar than more distant relatives in both physical features and DNA. This is because as cells divide DNA replicates. “The DNA makes two identical copies but occasionally there are mutations that change the sequence. Copies are thus mistakes that are then copied, and the mutations accumulate. You can count these differences in different organisms. This is one of the many methods to reconstruct evolutionary history. It all depends on the accumulation of mutations.”

Genes to words and more

Studying different sequences, developing the phylogenetic history and constructing evolutionary trees is therefore a routine part of Howe’s work. “This made us wonder about other things copied with the incorporation of changes, like ancient manuscripts copied by scribes. Although scribes tried to be accurate, mistakes or deletions got through. And textual scholars use the accumulated changes to work out the copying history – creating a stemma – a genealogical tree diagram that depicts the relationships between different manuscript versions of a particular text.” (The first stemma was created in 1827 and the approach was elaborated by Karl Lachmann around that time.) “These are basically asking the same question as we do when inferring the evolutionary history of organisms by looking at the similarities and differences between the DNA sequences. We therefore believed we could construct a stemma using the same computer programs used to map DNA evolutionary changes.” The process essentially involves turning each word into a DNA letter thus encoding the text as a DNA sequence and allowing a computer program to map the evolutionary tree based on where the words change. Howe and his colleagues used this technique initially on works like The Canterbury Tales, the poems of Robert Herrick and the Letter of James in the New Testament producing computer versions that correlate well with the manual stemma but were produced much faster. “Initially textual scholars were divided,” he said. “They were particularly concerned with contamination where you start with more than one version of a text, but biology has the same - genetic recombination or lateral gene transfer – bacteria do it all the time – where they transfer information to other bacteria. You then produce networks not trees. We found that using network methods when contaminated could be applied to textual data as well.” They have moved on to doing the same with music scores - specifically Bach’s Well-Tempered Clavier. “We found this more challenging to encode – Bach revised a lot himself and then others also copied.”  But the resulting version also correlated well with the manual work. Other scholars are using the same methods from evolutionary biology to study the development of languages, cultural artefacts and folktales. For languages you look at equivalent words across languages and use comparisons to draw a matrix of different languages to map their evolution. (During his stay at STIAS Howe has started some initial work on Khoisan languages.) There is also work that maps the evolution of patterns in Persian carpets. “This asks the question whether designs were passed by vertical descent (largely from mother to daughter) or from one village/group to another. “The tree-like structure produced indicated that direct linear transmission is probably a good model.” Similar methods have also been used to look at the development and spread of fairy and folktales. Howe’s group, in collaboration with researchers from the University of Cape Town, is now looking at how this method can track the evolution of a particular disease condition - Tuberous Sclerosis Complex, a rare genetic disorder that causes non-cancerous tumours to grow in various parts of the body affecting multiple systems and associated with symptoms ranging from skin and eye abnormalities to seizures, cognitive disabilities and kidney, lung and heart diseases. It’s hoped that evolutionary analysis can enhance understanding of the development of neuropsychological, behavioural and scholastic difficulties in such individuals. But Howe was quick to point out that this work is not a replacement for existing scholarship in any of these fields. “It provides a valuable set of tools to handle large amounts of data to answer questions in a wide range of areas. Tools that allow you to answer questions quickly and tell you where to focus further attention. It’s not a replacement for conventional scholarship. It relies on data from scholars and their analysis.” “Of course, it’s also not the right model for everything. Scholars have to interpret and there may be multiple interpretations,” he added. “You should never follow a computer program over an intellectual precipice.” “But copying with incorporation of changes is a fundamental feature of how Life works.” He ended by quoting physician, biologist and writer Lewis Thomas – “The capacity to blunder slightly is the real marvel of DNA. Without this special attribute we would still be anaerobic bacteria and there would be no music.”   Michelle Galloway: Part-time media officer at STIAS Photograph:  Noloyiso Mtembu