crc.pitt.edu

Clark and Chikina labs employ CRC in exploring convergent evolution

The star-nosed mole of North America and the naked mole rat of East Africa are both blind. They underwent the same adapatation to living underground, although they are different species separated by thousands of miles. Marine mammals like manatees and dolphins underwent shared adaptations to aquatic life. What could the convergence of independent physical changes reveal about the evolution of the genes responsible for those physical changes? The labs of Maria Chikina and Nathan Clark explore this evolution relying on CRC resources for computation tasks to compare rates of evolution for a gene in one species to rates of evolution for a gene in another species.

How Stars Explode

In the spring of 1006 CE a supernova in the constellation Lupus was the brightest stellar object ever recorded on Earth, bright enough for several months to be easily visible in daylight. Most supernovae are not dramatically visible from Earth and don’t leave visible evidence. What they do leave are supernova remnants: expanding balls of gas heated to millions of degrees Celsius. The remnants hold clues to the origins and deaths of stars, and the lab of Pitt astrophysicist Carles Badenes searches for those clues helped by the resources of the Center for Research Computing.

Stefania Albanesi returns to CRC to create a new model credit score

Two years ago Pitt economist Stefania Albanesi co-authored a paper refuting conventional wisdom that the 2007-2009 global credit crisis was triggered by mortgage defaults of borrowers with low credit scores. Albanesi found that borrowers with higher credit scores accounted for an outsized percentage of mortgage defaults. Albanesi and graduate student Domonkos Vamossy asked two questions – Why didn’t credit scores predict default particularly well? Can one create credit scores that pedict better? They developed a deep learning model to predict consumer default that outperforms standard credit scoring models in accuracy, a model that could help policy makers reduce consumer default and large-scale risks.

Building big data tools for ecology, forensics and more

Lucas Mentch uses CRC resources to explore the intersection of statistics and machine learning, creating models for a wide range of fields – ecology, criminal forensics, and sports analytics among them. One project worked with bird population data about the migration of tree swallows from the Cornell University Lab of Ornithology’s eBird project.

“eBird creates an amazing level of detail, so the migration presented many complex forms of local variation that contribute to a larger pattern,” Mentch explains. “The challenge was not only to build a model that produced accurate predictions, but also to develop testing procedures that would allow us to isolate the effects of individual variables.”

Tree swallow in flight (courtesy eBird)

Standing out among Greeks worldwide

Yanni Mpourmpakis, one of CRC's’s most prolific collaborators, won the Bodossaki Foundation Distinguished Young Scientist Award, among the highest scientific awards given in his native Greece, presented by Greek president Prokopios Pavlopoulos in Athens on June 19. Mpourmpakis, Bicentennial Alumni faculty fellow and associate professor of chemical and petroleum engineering, works with the Computer-Aided Nano and Energy Lab in developing computational methods and simulations to design nanomaterials – including work on safe storage of radioactive waste, formation of kidney stones, and creating building blocks for plastics. Left, Mpourmpakis with the Bodossaki Award plaque in the Greek Nationality Room in the Cathedral of Learning.

Photo courtesy Swanson School of Engineering

Search form

Clark and Chikina labs employ CRC in exploring convergent evolution

How Stars Explode

Stefania Albanesi returns to CRC to create a new model credit score

Building big data tools for ecology, forensics and more

Standing out among Greeks worldwide

News & Highlights

Connecting shared genes and shared traits across species

How stars explode

Deep Learning to Predict Default

Innovation grows in the Allegheny Mountains

Pages

System Status

User Alerts

Additional SMP nodes: short-term fix for busy cluster