Abstract
Cancer Research UK, creator of the popular Cell Slider citizen science game, will soon premiere a mobile phone game in which players identify genetic mutations associated with particular cancers.
While you stand in line to pick up a latte or catch a bus this fall, you'll also be able to help discover new treatments for cancer. In July, Cancer Research UK announced plans for GeneGame, a mobile phone game in which players identify genetic mutations associated with particular cancers. Set to debut in October, GeneGame is the latest gaming effort to tap into the collective intelligence of the public to tackle complex questions in cancer research.
“Games are built around motivation, around getting people to be engaged in a task,” says Andrew Su, PhD, of the Scripps Research Institute in La Jolla, CA. Although not directly involved in the Cancer Research UK initiative, Su has created numerous such “crowdsourcing” projects in biology. “If we can harness even a little bit of that, just a small percentage of the time people spend gaming, that would be a huge resource for cancer research.”
In the Cell Slider initiative, participants examine images of breast tissue like this one, determine whether there are any cancer cells, and, if so, count the number of cancer cells. Players are then asked to determine the proportion of yellow-stained cells and the brightness of the yellow, which corresponds to greater estrogen receptor levels.
In the Cell Slider initiative, participants examine images of breast tissue like this one, determine whether there are any cancer cells, and, if so, count the number of cancer cells. Players are then asked to determine the proportion of yellow-stained cells and the brightness of the yellow, which corresponds to greater estrogen receptor levels.
Like all cancer research centers, Cancer Research UK is accumulating massive amounts of genetic data thanks to next-generation sequencing. “We believe that buried in our data there are breakthroughs that are going to be cures for cancer,” says Amy Carton, citizen science lead for Cancer Research UK, “but to date, there isn't a machine or algorithm that is as good as a human being at analyzing that data, which has led to severe bottlenecks in data analysis.”
To help unplug that bottleneck, in October 2012 Cancer Research UK premiered a crowdsourcing project called Cell Slider. The online game asks participants to classify images of breast tissue by matching them to cancer cells displayed with irregular, yellow-stained nuclei or to normal cells with healthy pink nuclei. Scientists can then combine that analysis with patient outcomes to assess the effectiveness of different treatments.
Three months after the game's debut, players had analyzed a dataset of images that would have taken Cancer Research UK scientists 18 months, says Carton. By the end of July, more than 200,000 people had made more than 1.6 million classifications, which will be used in a research paper to be published later this year, she adds.
Cell Slider results have proven accurate thanks to multiple analyses of every image, says Carton. Each tumor sample is viewed by at least five people, so even if one person does not classify the sample correctly, other participants do.
Cancer Research UK is not the only cancer group taking advantage of crowdsourcing. In September 2012, Su's lab released a computer game called The Cure, which pits players against a computer opponent, “Barney,” to build the best five-card hand of genes, based on descriptions of the genes and the players' own knowledge of their function. The data gathered from the game—to date played 8,083 times by 920 players—have aided in predicting breast cancer survival based on gene information from tumor samples. The Cure is designed to be played by cancer biologists, but Su aspires to create more research-related games for the general public.
For more news on cancer research, visit Cancer Discovery online at http://CDnews.aacrjournals.org.
RSS Feeds