As complete genomes are sequenced and data pile up, more than ever before science needs bioinformatics engineers and their sophisticated software to make sense out of the myriads of chemical cascades in cells.
What proteins and metabolites are present in a cell and which genes are active? And: do certain proteins interact with each other? In modern molecular biology, scientists conduct all kinds of experiments to gather data that lead to better understandings of intracellular activities.
“Yet due to the complexity of the cell and limitations of the measuring techniques, these data are often noisy and incomplete”. says researcher Yunlei Li. “Dealing with these noisy data poses a major challenge.”
Li has faced that challenge. Until recently she worked as a PhD student at the Pattern Recognition and Bioinformatics department (faculty of Electrical Engineering, Mathematics and Computer Science). Next week she will defend her thesis, titled ‘Exploiting noisy and incomplete biological data for prediction and knowledge’.
Li developed a computational program with which she compared the metabolic networks of yeast (Saccharomyces cerevisae) with that of a bacterium (Escherichia coli). One of the metabolic pathways she looked at was glycolysis, which is the pathway that generates energy in almost all organisms by burning glucose.
“Throughout evolution this pathway has hardly changed because it’s so crucial for the functioning of organisms”. Li explains. “If you find differences here between species, this can either mean that you have stumbled upon the limitations of the measuring techniques that lead to false data or that you have found a very interesting evolutionary change.” Either way the computer model can help biologists conduct their research, Li believes.
Part of Li’s work consisted of chopping up the metabolic networks of both species into small building blocks of just several matching reactions. It may have taken scientists blood, sweat and tears to map all reactions (about a thousand for each of the two species) and create networks, but Li wanted to start with a clean slate.
Her algorithm searches for new possibilities to connect the pieces – as if it is playing with Lego bricks – and provides information about the likelihood of the connections and the similarities of the connected reactions.
According to Li, her work was one of the first attempts to conduct a systematic alignment of the full metabolic networks of multiple species.
More experiments are needed to validate the model, however, and that bothers Li somewhat: “I’d like to find out faster if it works or not and whether people will use it.”
Li has also decided to make a career change. She now works in the field of performance and risk assessment for a consultancy and software company, specialised in finance. If biologists don’t listen to her, hopefully bankers will.
Aan: Roos Stone (20, IO)
Ontvangen: Dinsdag 20 oktober 08.52 uur
Van: Nicky Castricum (22, IO)
EEN OF ANDERE SCHOBBEJAK HEEFT HET STUUR VAN MN FIETS GEJAT! ARGH!
Wanneer Nicky dinsdagochtend zijn fiets wil pakken, ontdekt hij dat zijn stuur is gejat. Per sms laat hij het zijn studiegenootje Roos weten. De hoofdletters schijnen een gewoonte te zijn als hij sms’t, overgehouden aan zijn vorige mobiel die nog nooit van kleine letters had gehoord.
Overigens was het ontvreemde stuurwiel niet zomaar een stuur. De fiets was namelijk voorzien van een heus autostuur. ‘Heel IO’ig vastgemaakt’, aldus Roos. “Met veel staaldraad, een beetje wiebelig, maar heel cool was het.”
Of de fiets inmiddels alweer een echt stuur heeft weet Roos niet. “De laatste keer dat ik hem zag stuurde hij met een boomtak, nog gevaarlijker!”
Comments are closed.