AI start-up Vicarious claims it has created software able to successfully solve CATPCHAs (completely automated public Turing tests to tell computers and humans apart). The system reportedly has a 90 to 99 percent accuracy rate, depending on the type of CAPTCHA used, whether it must recognize and then type in displayed letters, numbers, or else those with symbols whose shapes have been distorted and sometimes placed on distracting backgrounds. The company is not releasing many details about how the technology solves CAPTCHAs, but does say the system can be trained with moving and static images. One possible benefit of such technology, according to MIT’s Technology Review website, is in accelerating the digitization of degraded texts and improving optical character recognition. Vicarious plans to build a vision system as well as artificial intelligence that can be used in robotics, medical image analysis, and several other applications based on this technology. (Reuters)(Forbes)(Technology Review)
Saturday, November 9, 2013
Company Says its AI Algorithm Cracks CAPTCHAs
Tuesday, January 29, 2013
Solving puzzles without a picture: New algorithm assembles chromosomes from next generation sequencing data
The powerful DNA sequencing methods developed about 15 years ago, known as next generation sequencing (NGS) technologies, create thousands of short fragments. In species whose genetics has already been extensively studied, existing information can be used to organize and order the NGS fragments, rather like using a sketch of the complete picture as a guide to a jigsaw puzzle. But as genome scientists push into less-studied species, it becomes more difficult to finish the puzzle.
To solve this problem, a team led by Harris Lewin, distinguished professor of evolution and ecology and vice chancellor for research at the University of California, Davis and Jian Ma, assistant professor at the University of Illinois at Urbana-Champaign created a computer algorithm that uses the known chromosome organization of one or more known species and NGS information from a newly sequenced genome to create virtual chromosomes.
"We show for the first time that chromosomes can be assembled from NGS data without the aid of a preexisting genetic or physical map of the genome," Lewin said.
The new algorithm will be very useful for large-scale sequencing projects such as G10K, an effort to sequence 10,000 vertebrate genomes of which very few have a map, Lewin said.
"As we have shown previously, there is much to learn about phenotypic evolution from understanding how chromosomes are organized in one species relative to other species," he said.
The algorithm is called RACA (for reference-assisted chromosome assembly), co-developed by Jaebum Kim, now at Konkuk University, South Korea, and Denis Larkin of Aberystwyth University, Wales. Kim wrote the software tool which was evaluated using simulated data, standardized reference genome datasets as well as a primary NGS assembly of the newly sequenced Tibetan antelope genome generated by BGI (Shenzhen, China) in collaboration with Professor Ri-Li Ge at Qinghai University, China. Larkin led the experimental validation, in collaboration with scientists at BGI, proving that predictions of chromosome organization were highly accurate.
Ma said that the new RACA algorithm will perform even better as developing NGS technologies produce longer reads of DNA sequence.
"Even with what is expected from the newest generation of sequencers, complete chromosome assemblies will always be a difficult technical issue, especially for complex genomes. RACA predictions address this problem and can be incorporated into current NGS assembly pipelines," Ma said.
Additional coauthors on the paper are Qingle Cai, Asan, Yongfen Zhang, and Guojie Zhang, BGI-Shenzhen, China; Loretta Auvil and Boris Capitanu, University of Illinois Urbana-Champaign.
The work was supported by grants from the National Science Foundation, National Institutes of Health, U.S. Department of Agriculture, National Research Foundation of Korea, Polish Grid Infrastructure Project, National Basic Research Program of China, Program of International S&T Cooperation of China and the National Natural Science Foundation of China.
Share this story on Facebook, Twitter, and Google:Other social bookmarking and sharing tools:
Story Source:
The above story is reprinted from materials provided by University of California - Davis. The original article was written by Andy Fell.
Note: Materials may be edited for content and length. For further information, please contact the source cited above.
Journal Reference:
Jaebum Kim, Denis M. Larkin, Qingle Cai, Asan, Yongfen Zhang, Ri-Li Ge, Loretta Auvil, Boris Capitanu, Guojie Zhang, Harris A. Lewin, and Jian Ma. Reference-assisted chromosome assembly. PNAS, January 10, 2013 DOI: 10.1073/pnas.1220349110Note: If no author is given, the source is cited instead.
Disclaimer: Views expressed in this article do not necessarily reflect those of ScienceDaily or its staff.
Saturday, January 5, 2013
Predicting what topics will trend on Twitter: Algorithm offers new technique for analyzing data that fluctuate over time
At the Interdisciplinary Workshop on Information and Decision in Social Networks at MIT in November, Associate Professor Devavrat Shah and his student, Stanislav Nikolov, will present a new algorithm that can, with 95 percent accuracy, predict which topics will trend an average of an hour and a half before Twitter's algorithm puts them on the list -- and sometimes as much as four or five hours before.
The algorithm could be of great interest to Twitter, which could charge a premium for ads linked to popular topics, but it also represents a new approach to statistical analysis that could, in theory, apply to any quantity that varies over time: the duration of a bus ride, ticket sales for films, maybe even stock prices.
Like all machine-learning algorithms, Shah and Nikolov's needs to be "trained": it combs through data in a sample set -- in this case, data about topics that previously did and did not trend -- and tries to find meaningful patterns. What distinguishes it is that it's nonparametric, meaning that it makes no assumptions about the shape of patterns.
Let the data decide
In the standard approach to machine learning, Shah explains, researchers would posit a "model" -- a general hypothesis about the shape of the pattern whose specifics need to be inferred. "You'd say, 'Series of trending things … remain small for some time and then there is a step,'" says Shah, the Jamieson Career Development Associate Professor in the Department of Electrical Engineering and Computer Science. "This is a very simplistic model. Now, based on the data, you try to train for when the jump happens, and how much of a jump happens.
"The problem with this is, I don't know that things that trend have a step function," Shah explains. "There are a thousand things that could happen." So instead, he says, he and Nikolov "just let the data decide."
In particular, their algorithm compares changes over time in the number of tweets about each new topic to the changes over time of every sample in the training set. Samples whose statistics resemble those of the new topic are given more weight in predicting whether the new topic will trend or not. In effect, Shah explains, each sample "votes" on whether the new topic will trend, but some samples' votes count more than others'. The weighted votes are then combined, giving a probabilistic estimate of the likelihood that the new topic will trend.
In Shah and Nikolov's experiments, the training set consisted of data on 200 Twitter topics that did trend and 200 that didn't. In real time, they set their algorithm loose on live tweets, predicting trending with 95 percent accuracy and a 4 percent false-positive rate.
Shah predicts, however, that the system's accuracy will improve as the size of the training set increases. "The training sets are very small," he says, "but we still get strong results."
Keeping pace
Of course, the larger the training set, the greater the computational cost of executing Shah and Nikolov's algorithm. Indeed, Shah says, curbing computational complexity is the reason that machine-learning algorithms typically employ parametric models in the first place. "Our computation scales proportionately with the data," Shah says.
But on the Web, he adds, computational resources scale with the data, too: As Facebook or Google add customers, they also add servers. So his and Nikolov's algorithm is designed so that its execution can be split up among separate machines. "It is perfectly suited to the modern computational framework," Shah says.
In principle, Shah says, the new algorithm could be applied to any sequence of measurements performed at regular intervals. But the correlation between historical data and future events may not always be as clear cut as in the case of Twitter posts. Filtering out all the noise in the historical data might require such enormous training sets that the problem becomes computationally intractable even for a massively distributed program. But if the right subset of training data can be identified, Shah says, "It will work."
"People go to social-media sites to find out what's happening now," says Ashish Goel, an associate professor of management science at Stanford University and a member of Twitter's technical advisory board. "So in that sense, speeding up the process is something that is very useful." Of the MIT researchers' nonparametric approach, Goel says, "it's very creative to use the data itself to find out what trends look like. It's quite creative and quite timely and hopefully quite useful."
Share this story on Facebook, Twitter, and Google:Other social bookmarking and sharing tools:
Story Source:
The above story is reprinted from materials provided by Massachusetts Institute of Technology.
Note: Materials may be edited for content and length. For further information, please contact the source cited above.
Note: If no author is given, the source is cited instead.
Disclaimer: Views expressed in this article do not necessarily reflect those of ScienceDaily or its staff.
Wednesday, January 2, 2013
Leap forward in brain-controlled computer cursors: New algorithm greatly improves speed and accuracy
When a paralyzed person imagines moving a limb, cells in the part of the brain that controls movement still activate as if trying to make the immobile limb work again. Despite neurological injury or disease that has severed the pathway between brain and muscle, the region where the signals originate remains intact and functional.
In recent years, neuroscientists and neuroengineers working in prosthetics have begun to develop brain-implantable sensors that can measure signals from individual neurons, and after passing those signals through a mathematical decode algorithm, can use them to control computer cursors with thoughts. The work is part of a field known as neural prosthetics.
A team of Stanford researchers have now developed an algorithm, known as ReFIT, that vastly improves the speed and accuracy of neural prosthetics that control computer cursors. The results are to be published Nov. 18 in the journal Nature Neuroscience in a paper by Krishna Shenoy, a professor of electrical engineering, bioengineering and neurobiology at Stanford, and a team led by research associate Dr. Vikash Gilja and bioengineering doctoral candidate Paul Nuyujukian.
In side-by-side demonstrations with rhesus monkeys, cursors controlled by the ReFIT algorithm doubled the performance of existing systems and approached performance of the real arm. Better yet, more than four years after implantation, the new system is still going strong, while previous systems have seen a steady decline in performance over time.
"These findings could lead to greatly improved prosthetic system performance and robustness in paralyzed people, which we are actively pursuing as part of the FDA Phase-I BrainGate2 clinical trial here at Stanford," said Shenoy.
Sensing mental movement in real time
The system relies on a silicon chip implanted into the brain, which records "action potentials" in neural activity from an array of electrode sensors and sends data to a computer. The frequency with which action potentials are generated provides the computer key information about the direction and speed of the user's intended movement.
The ReFIT algorithm that decodes these signals represents a departure from earlier models. In most neural prosthetics research, scientists have recorded brain activity while the subject moves or imagines moving an arm, analyzing the data after the fact. "Quite a bit of the work in neural prosthetics has focused on this sort of offline reconstruction," said Gilja, the first author of the paper.
The Stanford team wanted to understand how the system worked "online," under closed-loop control conditions in which the computer analyzes and implements visual feedback gathered in real time as the monkey neurally controls the cursor to toward an onscreen target.
The system is able to make adjustments on the fly when while guiding the cursor to a target, just as a hand and eye would work in tandem to move a mouse-cursor onto an icon on a computer desktop. If the cursor were straying too far to the left, for instance, the user likely adjusts their imagined movements to redirect the cursor to the right. The team designed the system to learn from the user's corrective movements, allowing the cursor to move more precisely than it could in earlier prosthetics.
To test the new system, the team gave monkeys the task of mentally directing a cursor to a target -- an onscreen dot -- and holding the cursor there for half a second. ReFIT performed vastly better than previous technology in terms of both speed and accuracy. The path of the cursor from the starting point to the target was straighter and it reached the target twice as quickly as earlier systems, achieving 75 to 85 percent of the speed of real arms.
"This paper reports very exciting innovations in closed-loop decoding for brain-machine interfaces. These innovations should lead to a significant boost in the control of neuroprosthetic devices and increase the clinical viability of this technology," said Jose Carmena, associate professor of electrical engineering and neuroscience at the University of California Berkeley.
A smarter algorithm
Critical to ReFIT's time-to-target improvement was its superior ability to stop the cursor. While the old model's cursor reached the target almost as fast as ReFIT, it often overshot the destination, requiring additional time and multiple passes to hold the target.
The key to this efficiency was in the step-by-step calculation that transforms electrical signals from the brain into movements of the cursor onscreen. The team had a unique way of "training" the algorithm about movement. When the monkey used his real arm to move the cursor, the computer used signals from the implant to match the arm movements with neural activity. Next, the monkey simply thought about moving the cursor, and the computer translated that neural activity into onscreen movement of the cursor. The team then used the monkey's brain activity to refine their algorithm, increasing its accuracy.
The team introduced a second innovation in the way ReFIT encodes information about the position and velocity of the cursor. Gilja said that previous algorithms could interpret neural signals about either the cursor's position or its velocity, but not both at once. ReFIT can do both, resulting in faster, cleaner movements of the cursor
An engineering eye
Early research in neural prosthetics had the goal of understanding the brain and its systems more thoroughly, Gilja said, but he and his team wanted to build on this approach by taking a more pragmatic engineering perspective. "The core engineering goal is to achieve highest possible performance and robustness for a potential clinical device, " he said.
To create such a responsive system, the team decided to abandon one of the traditional methods in neural prosthetics. Much of the existing research in this field has focused on differentiating among individual neurons in the brain. Importantly, such a detailed approach has allowed neuroscientists to create a detailed understanding of the individual neurons that control arm movement.
The individual neuron approach has its drawbacks, Gilja said. "From an engineering perspective, the process of isolating single neurons is difficult, due to minute physical movements between the electrode and nearby neurons, making it error-prone," he said. ReFIT focuses on small groups of neurons instead of single neurons.
By abandoning the single-neuron approach, the team also reaped a surprising benefit: performance longevity. Neural implant systems that are fine-tuned to specific neurons degrade over time. It is a common belief in the field that after six months to a year, they can no longer accurately interpret the brain's intended movement. Gilja said the Stanford system is working very well more than four years later.
"Despite great progress in brain-computer interfaces to control the movement of devices such as prosthetic limbs, we've been left so far with halting, jerky, Etch-a-Sketch-like movements. Dr. Shenoy's study is a big step toward clinically useful brain-machine technology that have faster, smoother, more natural movements," said James Gnadt, PhD, a program director in Systems and Cognitive Neuroscience at the National Institute of Neurological Disorders and Stroke, part of the National Institutes of Health.
For the time being, the team has been focused on improving cursor movement rather than the creation of robotic limbs, but that is not out of the question, Gilja said. Near term, precise, accurate control of a cursor is a simplified task with enormous value for paralyzed people.
"We think we have a good chance of giving them something very useful," he said. The team is now translating these innovations to paralyzed people as part of a clinical trial.
This research was funded by the Christopher and Dana Reeve Paralysis Foundation; NSF, NDSEG, and SGF Graduate Fellowships; DARPA ("Revolutionizing Prosthetics" and "REPAIR"); and NIH (NINDS-CRCNS and Director's Pioneer Award).
Other contributing researchers include Cynthia Chestek, John Cunningham, and Byron Yu, Joline Fan, Mark Churchland, Matthew Kaufman, Jonathan Kao, and Stephen Ryu.
Share this story on Facebook, Twitter, and Google:Other social bookmarking and sharing tools:
Story Source:
The above story is reprinted from materials provided by Stanford School of Engineering. The original article was written by Kelly Servick, science-writing intern.
Note: Materials may be edited for content and length. For further information, please contact the source cited above.
Journal Reference:
Vikash Gilja, Paul Nuyujukian, Cindy A Chestek, John P Cunningham, Byron M Yu, Joline M Fan, Mark M Churchland, Matthew T Kaufman, Jonathan C Kao, Stephen I Ryu, Krishna V Shenoy. A high-performance neural prosthesis enabled by control algorithm design. Nature Neuroscience, 2012; DOI: 10.1038/nn.3265Note: If no author is given, the source is cited instead.
Disclaimer: This article is not intended to provide medical advice, diagnosis or treatment. Views expressed here do not necessarily reflect those of ScienceDaily or its staff.
Wednesday, October 31, 2012
Fast algorithm extracts and compares document meaning
Writing in the International Journal of Intelligent Information and Database Systems, Andrzej Sieminski of the Technical University of Wroclaw, explains that extracting meaning and calculating the level of semantic similarity between two pieces of texts is a very difficult task, without human intervention. There have been various methods proposed by computer scientists for addressing this problem, but they all suffer from computational complexity, he says.
Sieminski has now attempted to reduce this complexity by merging a computationally efficient statistical approach to text analysis with a semantic component. Tests of the algorithm on English and Polish tests work well. The test set consisted of 4,890 English sentences with 142,116 words and 11,760 Polish sentences with 184,524 words scraped from online services via their newsfeeds over the course of five days. Sieminski points out that the complexity of the algorithm used on the Polish documents required an additional level of sophistication in terms of computing word means and disambiguation.
Traditional "manual" methods of indexing simply cannot now cope with the vast quantities of information generated on a daily basis by humanity as a whole in scientific research more specifically. The new algorithm once optimised could radically change the way in which we make archived documents searchable and allow knowledge to be extracted far more readily than is possible with standard indexing and search tools.
The approach also circumvents three critical problems faced by most users of conventional search engines: First, the lack of familiarity with the advanced search options of search engines, with a semantic algorithm advanced options become almost unnecessary. Secondly, the rigid nature of the options that are unable to catch the subtle nuance of user information needs, again a tool that understands the meaning of a search and the meaning of the results it offers avoids this problem. Finally, the unwillingness or unacceptably long time necessary to type a long query, semantically aware search will require only simply input.
Sieminski points out that the key virtue of the research is the idea of using the statistical similarity measures to assess semantic similarity. He explains that semantic similarity of words could be inferred from the WordNet database. He proposes using this database only during text indexing. "Indexing is done only once so the inevitably long processing time is not an issue," he says. "From that point on we use only statistical algorithms, which are fast and high performance."
Share this story on Facebook, Twitter, and Google:Other social bookmarking and sharing tools:
Story Source:
The above story is reprinted from materials provided by Inderscience, via AlphaGalileo.
Note: Materials may be edited for content and length. For further information, please contact the source cited above.
Journal Reference:
Andrzej Sieminski. Fast algorithm for assessing semantic similarity of texts. Int. J. Intelligent Information and Database Systems, 2012, 6, 495-512Note: If no author is given, the source is cited instead.
Disclaimer: Views expressed in this article do not necessarily reflect those of ScienceDaily or its staff.
Thursday, October 18, 2012
Computer, read my lips: Emotion detector developed using a genetic algorithm
Karthigayan Muthukaruppanof Manipal International University in Selangor, Malaysia, and co-workers have developed a system using a genetic algorithm that gets better and better with each iteration to match irregular ellipse fitting equations to the shape of the human mouth displaying different emotions. They have used photos of individuals from South-East Asia and Japan to train a computer to recognize the six commonly accepted human emotions -- happiness, sadness, fear, angry, disgust, surprise -- and a neutral expression. The upper and lower lip is each analyzed as two separate ellipses by the algorithm.
"In recent years, there has been a growing interest in improving all aspects of interaction between humans and computers especially in the area of human emotion recognition by observing facial expression," the team explains. Earlier researchers have developed an understanding that allows emotion to be recreated by manipulating a representation of the human face on a computer screen. Such research is currently informing the development of more realistic animated actors and even the behavior of robots. However, the inverse process in which a computer recognizes the emotion behind a real human face is still a difficult problem to tackle.
It is well known that many deeper emotions are betrayed by more than movements of the mouth. A genuine smile for instance involves flexing of muscles around the eyes and eyebrow movements are almost universally essential to the subconscious interpretation of a person's feelings. However, the lips remain a crucial part of the outward expression of emotion. The team's algorithm can successfully classify the seven emotions and a neutral expression described.
The researchers suggest that initial applications of such an emotion detector might be helping disabled patients lacking speech to interact more effectively with computer-based communication devices, for instance.
Share this story on Facebook, Twitter, and Google:Other social bookmarking and sharing tools:
Story Source:
The above story is reprinted from materials provided by Inderscience Publishers, via EurekAlert!, a service of AAAS.
Note: Materials may be edited for content and length. For further information, please contact the source cited above.
Journal Reference:
M. Karthigayan, R. Nagarajan, M. Rizon, Sazali Yaacob. Lip pattern in the interpretation of human emotions. International Journal of Artificial Intelligence and Soft Computing, 2012; 3 (2): 95 DOI: 10.1504/IJAISC.2012.049004Note: If no author is given, the source is cited instead.
Disclaimer: This article is not intended to provide medical advice, diagnosis or treatment. Views expressed here do not necessarily reflect those of ScienceDaily or its staff.
Saturday, September 22, 2012
Math tree may help root out fraudsters: Applying algorithm to social networks can reveal hidden connections criminals use to commit fraud
The Steiner tree, that is.
In an article recently published in the journal Computer Fraud and Security, University of Alberta researcher Ray Patterson and colleagues from the University of Connecticut and University of California -- Merced outlined the connection linking fraud cases and the algorithm designed by Swiss mathematician Jakob Steiner. Fraud is a problem that costs Canadians billions of dollars annually and countless hours of police investigations. Patterson says that building the algorithm into fraud investigation software may provide important strategic advantages.
The criminal path of least resistance
To quote a television gumshoe, everything's connected. Figuring out who knows who and who has access to the money is like playing a game of connect-the-dots. Patterson says that for crimes like fraud, the fewer players in the scheme, the more likely it will be accomplished. Maintaining a small group of players is also what links it to the Steiner tree. He says that by analyzing various connecting social networks -- email, Facebook or the like -- finding out the who, what and how of the crime can be boiled down to numbers.
"You're really trying to find the minimum set of connectors that connect these people to the various [network] resources," he said. "The minimum number of people required is what's most likely to be the smoking gun. You can do it with math, once you know what the networks are."
Fraud and the Steiner tree, by the numbers
In their article, Patterson and his colleagues explored how networks such as phone calls, business partnerships and family relationships are used to form essential relationships in a fraud investigation. When these same relationships are layered, a pattern of connection becomes obvious. Once unnecessary links are removed and false leads are extracted, the remaining connections are most likely the best suspects. Patterson says that finding the shortest connection between the criminals and the crime is the crux of the Steiner tree.
"All of these things that we see in life, behind them is a mathematical representation," said Patterson. "There are many, many different algorithms that we can pull off a shelf and apply to real-life problems."
A potential tool for the long arm of the law?
Patterson says that with the amount of work that could potentially go into investigating a fraud case, such as obtaining warrants for phone or email records, and identifying and interviewing potential suspects, developing a program that uses a Steiner tree algorithm may save a significant portion of investigators' time -- time that, he says, could likely be reallocated to backlog or cold case files. "If you can reduce your legwork by even 20 per cent, that has massive manpower implications. I think algorithms like this one could help you reduce your legwork a lot more than that," he said.
Although there is software that police and other law enforcement agencies can use to solve fraud, Patterson sees no evidence that those programs use a Steiner tree algorithm, something he says would bring some structure to an unstructured area. He hopes programmers and investigators will take note of the findings and make changes to their practices.
"It might take several years or many years before anyone picks it up," said Patterson. "But it's a good thing if we can point people towards what's useful."
Share this story on Facebook, Twitter, and Google:Other social bookmarking and sharing tools:
Story Source:
The above story is reprinted from materials provided by University of Alberta, via EurekAlert!, a service of AAAS.
Note: Materials may be edited for content and length. For further information, please contact the source cited above.
Journal Reference:
Ram D Gopal, Raymond A Patterson, Erik Rolland, Dmitry Zhdanov. Social network meets Sherlock Holmes: investigating the missing links of fraud. Computer Fraud & Security, 2012; 2012 (7): 12 DOI: 10.1016/S1361-3723(12)70074-XNote: If no author is given, the source is cited instead.
Disclaimer: Views expressed in this article do not necessarily reflect those of ScienceDaily or its staff.
Friday, August 31, 2012
Frog calls inspire a new algorithm for wireless networks
How can network nodes be coloured with the least possible number of colours without two consecutive nodes being the same colour? A team of researchers at the Polytechnic University of Catalonia have found a solution to this mathematical problem with the help of some rather special colleagues: Japanese tree frogs (Hyla japonica).
These male amphibians use their calls to attract the female, who can recognise where it comes from and then locate the suitor. The problem arises when two males are too close to one another and they use their call at the same time. The females become confused and are unable to determine the location of the call. Therefore, the males have had to learn how to 'desynchronise' their calls or, in other words, not call at the same time in order for a distinction to be made.
"Since there is no system of central control organising this "desynchronisation," the mechanism may be considered as an example of natural self-organisation," explains Christian Blum. With the help of his colleague Hugo Hernández, such behaviour provided inspiration for "solving the so-called 'graph colouring problem' in an even and distributed way."
A graph is a set of connected nodes. As in the case of the frog's 'desynchronised calls', operating in a 'distributed' fashion implies that there is no other way of central control that helps to solve the problem with a global vision and all the information on the situation.
In the same way, the researchers have devised a new algorithm for assigning colours to network nodes ensuring that each pair of connected nodes is not the same colour. The end goal is to generate a valid solution that uses the least amount of colours.
Application to WiFi connections
As Blum outlines, "this type of graph colouring is the formalisation of a problem that arises in many areas of the real world, such as the optimisation of modern wireless networks with no predetermined structure using techniques for reducing losses in information packages and energy efficiency improvement."
This study falls under the field of 'swarm intelligence', a branch of artificial intelligence that aims to design intelligent systems with multiple agents. This is inspired by the collective behaviour of animal societies such as ant colonies, flocks of birds, shoals of fish and frogs, as in this case.
Share this story on Facebook, Twitter, and Google:Other social bookmarking and sharing tools:
Story Source:
The above story is reprinted from materials provided by Plataforma SINC, via AlphaGalileo.
Note: Materials may be edited for content and length. For further information, please contact the source cited above.
Journal Reference:
Hugo Hernández, Christian Blum. Distributed graph coloring: an approach based on the calling behavior of Japanese tree frogs. Swarm Intelligence, 2012; 6 (2): 117 DOI: 10.1007/s11721-012-0067-2Note: If no author is given, the source is cited instead.
Disclaimer: Views expressed in this article do not necessarily reflect those of ScienceDaily or its staff.
Sunday, August 26, 2012
Google goes cancer: Search engine algorithm finds cancer biomarkers
The researcher's own version of the Google algorithm has been used in this study to find new cancer biomarkers, which are molecules produced by cancer cells. Biomarkers can help to detect cancer earlier in body fluids or directly in the cancer tissue obtained in an operation or biopsy. Finding these biomarkers is often difficult and time consuming. Another problem is that markers found in different studies for the same types of cancer almost never overlap.
This problem has been circumvented using the Google strategy, which takes into account the content of a web page and also how these pages are connected via hyperlinks. With this strategy as the model, the authors made use of the fact that proteins in a cell are connected through a network of physical and regulatory interactions; the 'protein Facebook' so to speak.
"Once we added the network information in our analysis, our biomarkers became more reproducible," said Christof Winter, the paper's first author. Using this network information and the Google Algorithm, a significant overlap was found with an earlier study from the University of North Carolina. There, a connection was made with a protein which can assess aggressiveness in pancreatic cancer.
Although the new biomarkers seem to mark an improvement over currently used diagnostic tools, they are far from perfect and still need to be validated in a larger follow-up study before they can be used in clinical practice. It remains an open problem to turn these insights into novel drugs which slow down cancer progression. A first step in this direction is the group's cooperation with the Dresden-based biotech company RESprotect, who are running a clinical trial on a pancreas cancer drug.
TU Dresden is a leading German university, whose Center for Regenerative Therapies was awarded excellence status in the national excellence initiative. The work was a cooperation between the bioinformatics group of Prof. Dr. Michael Schroeder and the medical groups of Dr. Christian Pilarsky and Prof. Robert Grützmann.
Share this story on Facebook, Twitter, and Google:Other social bookmarking and sharing tools:
Story Source:
The above story is reprinted from materials provided by Public Library of Science.
Note: Materials may be edited for content and length. For further information, please contact the source cited above.
Journal Reference:
Christof Winter, Glen Kristiansen, Stephan Kersting, Janine Roy, Daniela Aust, Thomas Knösel, Petra Rümmele, Beatrix Jahnke, Vera Hentrich, Felix Rückert, Marco Niedergethmann, Wilko Weichert, Marcus Bahra, Hans J. Schlitt, Utz Settmacher, Helmut Friess, Markus Büchler, Hans-Detlev Saeger, Michael Schroeder, Christian Pilarsky, Robert Grützmann. Google Goes Cancer: Improving Outcome Prediction for Cancer Patients by Network-Based Ranking of Marker Genes. PLoS Computational Biology, 2012; 8 (5): e1002511 DOI: 10.1371/journal.pcbi.1002511Note: If no author is given, the source is cited instead.
Disclaimer: This article is not intended to provide medical advice, diagnosis or treatment. Views expressed here do not necessarily reflect those of ScienceDaily or its staff.
Saturday, June 2, 2012
Google goes cancer: Search engine algorithm finds cancer biomarkers
The researcher's own version of the Google algorithm has been used in this study to find new cancer biomarkers, which are molecules produced by cancer cells. Biomarkers can help to detect cancer earlier in body fluids or directly in the cancer tissue obtained in an operation or biopsy. Finding these biomarkers is often difficult and time consuming. Another problem is that markers found in different studies for the same types of cancer almost never overlap.
This problem has been circumvented using the Google strategy, which takes into account the content of a web page and also how these pages are connected via hyperlinks. With this strategy as the model, the authors made use of the fact that proteins in a cell are connected through a network of physical and regulatory interactions; the 'protein Facebook' so to speak.
"Once we added the network information in our analysis, our biomarkers became more reproducible," said Christof Winter, the paper's first author. Using this network information and the Google Algorithm, a significant overlap was found with an earlier study from the University of North Carolina. There, a connection was made with a protein which can assess aggressiveness in pancreatic cancer.
Although the new biomarkers seem to mark an improvement over currently used diagnostic tools, they are far from perfect and still need to be validated in a larger follow-up study before they can be used in clinical practice. It remains an open problem to turn these insights into novel drugs which slow down cancer progression. A first step in this direction is the group's cooperation with the Dresden-based biotech company RESprotect, who are running a clinical trial on a pancreas cancer drug.
TU Dresden is a leading German university, whose Center for Regenerative Therapies was awarded excellence status in the national excellence initiative. The work was a cooperation between the bioinformatics group of Prof. Dr. Michael Schroeder and the medical groups of Dr. Christian Pilarsky and Prof. Robert Grützmann.
Share this story on Facebook, Twitter, and Google:Other social bookmarking and sharing tools:
Story Source:
The above story is reprinted from materials provided by Public Library of Science.
Note: Materials may be edited for content and length. For further information, please contact the source cited above.
Journal Reference:
Christof Winter, Glen Kristiansen, Stephan Kersting, Janine Roy, Daniela Aust, Thomas Knösel, Petra Rümmele, Beatrix Jahnke, Vera Hentrich, Felix Rückert, Marco Niedergethmann, Wilko Weichert, Marcus Bahra, Hans J. Schlitt, Utz Settmacher, Helmut Friess, Markus Büchler, Hans-Detlev Saeger, Michael Schroeder, Christian Pilarsky, Robert Grützmann. Google Goes Cancer: Improving Outcome Prediction for Cancer Patients by Network-Based Ranking of Marker Genes. PLoS Computational Biology, 2012; 8 (5): e1002511 DOI: 10.1371/journal.pcbi.1002511Note: If no author is given, the source is cited instead.
Disclaimer: This article is not intended to provide medical advice, diagnosis or treatment. Views expressed here do not necessarily reflect those of ScienceDaily or its staff.