An online game has helped scientists from McGill University, Montreal, trace the origins of 521 genetic diseases over the past year. Using humans’ abilities to recognise and sort patterns, the game has significantly advanced understanding of the genetic basis of diseases such as Alzheimer’s, diabetes and cancer.
Game designer and assistant professor in the School of Computer Science, Jérôme Waldispühl, wanted to “recycle the energy that people spend on computer games into something useful”, so he developed Phylo.
Serious games
Phylo, launched in November 2010, enables users to sort genetic code to find similarities between regions of DNA in a process called multiple sequence alignment. Users of the game arrange two or more lines of coloured squares that represent DNA sequences in order to spot genetic relationships between animal species and hereditary diseases.
When two portions of DNA are compared, the differences within target areas highlight where a mutation on a gene causes a certain disease. This helps researchers track the evolution of the illness and tailor drugs to specific markers on the genetic code.
In order to find the optimal arrangement of DNA, a computer algorithm would have to explore all possible solutions one by one. With an average of one billion DNA blocks to sort per genome, the computation would take forever.
Phylo encourages the unscientific method of guesswork, harnessing the human mind’s aptitude for working out visual puzzles. Adding an element of competition and reward ensures gamers return to play again.
“Instead of looking at all possible alignments and ways to arrange the different sequences of colour blocks… we used human intuition to find the best alignment just by guessing what the configuration should be,” says WaldispĂĽhl.
The data for the game is sourced from the University of California Santa Cruz (UCSC) Genome Browser, where scientists publicly release their sequence alignments. These sequences have been prealigned with a programme that uses heuristics, a trial-and-error computation method that results in misaligned regions of sequence.
Waldispühl explains this synergy between computer and humans, “computers do the bulk of the job at a lower level of resolution and when it gets more precise, we rely more on humans”.
Game screen for Phylo game
DNA Tetris
The process of gamification – transforming repetitive tasks into interactive challenges with a reward system – is an emerging field in the biology world. Users of a protein folding game called Foldit solved the structure for an AIDs causing protein in rhesus monkeys that had previously remained unsolved for 15 years. The result was confirmed by x-ray crystallography.
Proteins are in the form of a random coil after sequencing, but a correct three-dimensional structure is essential for the compounds to function. Failure to fold correctly produces proteins that are toxic.
Foldit uses the classical approach of trying to do a job that a computer cannot do well, but in the process users have to learn the science behind how a protein folds. This is where Phylo is different.
“Phylo is quite novel,” says Waldispühl, “We tried to abstract the science and create a casual game.” The result is a Tetris-like puzzle that Phylo’s 17,000 online users can play in the background whilst browsing the web with the knowledge that their actions help make better genetic computations.
Citizen science
This isn’t the first time genetics and crowdsourcing have gone hand in hand. One of the earliest landmark examples of crowdsourcing was GenBank, launched in 1982 as a depository for genetic researchers. Essentially an open access database for any new genome sequences, the project started with 2,000 entries and doubled submissions every 18 months.
The cost of genome sequencing, although falling rapidly since the $3 billion per genome in the eighties, is still prohibitively expensive on a large scale. By compiling their research, scientists were able to cut time and cost and so massively advance the field of genetics.
It looks like channelling small amounts of energy from a large number of people and entwining computer and human strengths is the future of genetic discovery.
In the words of Phylo: Thank you for contributing to science.
http://dx.doi.org/10.1038/nsmb.2119
http://dx.doi.org/10.1093/nar/gkm929