In this edition of Research Connections, find links to researchers in the news, updates on important deadlines, and more news for University of Rochester researchers. Email not displaying correctly?
View it in your browser.

At upper left, a healthy astrocyte — a supportive brain cell — is shown in blue between green sheaths of myelin, which are produced by oligodendrocytes, the tentacled objects also seen in green. In individuals suffering from Progressive multifocal leukoencephalopathy, JC viruses (red particles) first infect the astrocyte (upper right) and mutate, eventually causing the astrocyte to explode (bottom image). The viruses then infect the oligodendrocytes.

Mouse model helps researchers target deadly brain disease

When University researchers Steven Goldman and Maiken Nedergaard created a mouse model whose brains consisted of both animal neurons and human glia cells, their study initially focused on findings that the human cells essentially made the mice smarter.

However, they also created a powerful new platform for researchers to study human glial cells in experimental animals. And that is providing new insights into Progressive multifocal leukoencephalopathy (PML).

This disease of the brain's white matter is caused by a virus infection that destroys the oligodendrocyte cells that make myelin — the fatty material that insulates nerve fibers. The most prominent symptoms are clumsiness; progressive weakness; and visual, speech, and personality changes.

The progression of the disease leads to life-threatening disability and often death, all due to its causative virus, the polyomavirus JC virus. JCV is common in humans, and is harmless to most, except to those whose immune systems are suppressed, whether as a result of chemotherapy, or HIV infection, or from treatment with immuno-suppressants for neurological disorders. Moreover, because the virus attacks only human glia brain cells — but not the glia cells in mice or other animals commonly used in experiments — it has been difficult to study the actual mechanism of the disease.

The new mouse model has allowed Goldman and his team — including Martha Windrem, Assistant Professor of Neurology who helped pioneer both the chimeric mouse model and its use in modeling JCV — to observe the effects of JCV in the live brain. To do so, they inject baby mice with human glial progenitor cells, which then migrate through the brain, giving rise to both astrocytes and oligodendrocytes. The researchers then inject JC virus, and are able to to track the impact of the infection as it unfolds in real time. They observed that the initial target of the virus is, in fact, astrocytes and, to a lesser extent, glial progenitor cells. The astrocytes serve as hosts for the virus to replicate and mutate, to the point where the cells literally explode and spread the infection in a chain reaction-like pattern (see illustration above). This, in turn, causes the oligodendrocytes to become infected.

"Now that we have focused on the astrocytes as a target for therapeutic development — because we know the astrocyte is a primary target for the virus itself — we can go after preventing viral infection of the astrocytes, and preventing viral replication in the astrocytes, as our treatment strategies," Goldman explains in a video. "We think we now have a tremendous opportunity for going after the infection of the brain by the JC virus in ways we have not been able to previously, because we simply did not have that degree of cellular understanding of the process by which the virus is infecting the brain and wreaking the havoc that it does." Read more . . .

Do you have an interesting photo or other image that helps illustrate your research? We would like to showcase it. Send a high resolution jpg or other version, along with a description of what it shows, to

Money, coverage contribute to "negative" campaigns

(One in a series about Simon School researchers and their use of data science.)

The closer a political campaign, the more likely it is that the candidates will attack each other with negative ads. This correlation has often been cited, and seems straightforward enough.

However, when Mitchell J. Lovett, Associate Professor of Marketing at the Simon School, and collaborator Ron Shachar took a closer look at what actually motivates candidates to "go negative," they discovered that two other factors are even more important than the closeness of the race. Specifically, negative ads increase as voters become more knowledgeable about the candidates — through the candidates' own ads and through media coverage — and when candidates have larger amounts of money to spend.

Lovett and Shachar undertook their study, "Seeds of Negativity: Knowledge and Money," because most research on negative advertising — in both politics and commerce — focuses on consumer (voter) reactions to it. They wanted to create a model to help explain why candidates choose to go negative — and also to account for the fact that hotly contested, "negative" campaigns also tend to have higher campaign spending and more media coverage, which increases voter knowledge of the candidates. These factors had been largely ignored in previous studies on the motivations for "going" negative.

Lovett and Shachar turned to psychology literature for some of the assumptions used in their model, specifically that "bad" information is processed more thoroughly than good, as embodied in a 2001 Review of General Psychology article entitled "Bad is Stronger than Good." They also drew on research showing that people process negative and positive messages through "dual accounts" — much as they might compile a list of pros and cons. "Negative and positive messages don't cancel in a one-to-one way," Lovett noted, "and, after receiving enough messages, another negative message will likely influence you more than another positive message."

Lovett and Shachar used data from all U.S. Congressional races in the top 75 Nielsen designated media areas in 2000, and all Congressional races in the top 100 such areas in 2002 and 2004. Included in this time frame was a Congressional vote that received considerable media attention, authorizing use of military force against Iraq.

"One of the nice things about studying political races is that there are lots of them," Lovett noted. "Also, they're held in similar settings. And they end. You have a clear winner and loser. You don't often have that in commercial settings."

Even so, he believes the model developed in this study could be useful in analyzing negative product advertising by commercial companies — especially among well-known, established competitors, and especially when there are economic downturns.

As Lovett and Shachar note in their study: "when the economy is not growing, the only way that a firm can grow is at the expense of its competitors. This means greater competition for the same consumers, and the parallel to political races is fairly immediate."

Click here to read the study.

Researchers train computers to accurately assess emotions elicited by images

Jiebo Luo, Professor of Computer Science; PhD student Quanzeng You; and researchers at Adobe Research have come up with a more accurate way to train computers to digest data that comes in the form of images. In a paper presented at the American Association for Artificial Intelligence (AAAI) conference in Austin, Texas, they describe a progressive training deep convolutional neural network (CNN). The trained computer can then be used to determine what sentiments these images are likely to elicit.

Luo says that this information could be useful for things as diverse as measuring economic indicators or predicting elections. Read more . . .

To begin the training process, Luo and his collaborators used a huge number of Flickr images that have been loosely labeled with specific sentiments by a machine algorithm. This gives the computer a starting point to begin understanding what some images can convey. But the machine-generated labels also include a likelihood of that label being true. The key step of the training process involves discarding any images for which the labels might not be true. Using only the "better" labeled images for further training significantly improved the accuracy of the sentiments with which each picture is labeled.

The paper can be downloaded here.

Visiting scholar will address Big Data and the Humanities

Laura Mandell, Director of the Initiative for Digital Humanities, Media, and Culture at Texas A&M University, will be here Feb. 19-20 as the University's first-ever Visiting Scholar in Digital Humanities. Her research focuses on visualizing poetry, developing software that will allow all scholars to deep-code documents for data-mining, and improving OCR software for early modern and 18th century texts via high performance and cluster computing.

Her visit, supported by the Andrew W. Mellon Fellows in Digital Humanities, will include these events, which are open to the public:

Thursday, Feb. 19:
Presentation: "Big Data and the Humanities," 10:30 a.m., VISTA Collaboratory.
Lunch with graduate students, 12:30 p.m., Digital Humanities Center.
Keynote: "Scaling Up: Search as Research," 5 p.m., Morey 321.

Friday, Feb. 20:
Panel: "Archives in Between: Cultural Preservation, Material to Digital," 10:30 a.m., Welles-Brown Room, with Joanne Bernardi, Associate Professor of Japanese and Film & Media Studies, and author/editor of ReEnvisioning Japan; Daniela Currò, Preservation Manager, Moving Image Department, George Eastman House; and Jim Kuhn, Joseph N. Lambert and Harold B. Schleifer Director of Rare Books and Special Collections.
DH Lunch: "The Dark Side of Digital Humanities," 12:30 p.m. at the Welles-Brown Room. For this event, RSVP to by end of today.

Eastman celebrates Frescobaldi publication with lecture, master classes

Girolamo Frescobaldi's "First Book of Toccatas," published in 1615, opened a new style of toccata composition whose basic elements are the "contrasting affects." The composition was no longer built on the traditional rules of counterpoint but, as Frescobaldi says in his preface, on the emotions, like in the madrigal. Rhetoric, not counterpoint, became the interpretive key. This change made the toccata a journey through the emotions, a story told at the keyboard where, in the absence of text, musical figures were entrusted to communicate emotion to the listener.

This publication's fame and success ensured that it spread quickly throughout Europe, influencing many major composers.

Two Eastman School of Music departments are collaborating on a five-event celebration of the 400th anniversary of Frescobaldi's seminal work, including a lecture, master classes, and performances of the composer's seminal work.

The celebration starts at 7 p.m., March 2, with a lecture, "From Madrigal to Toccata," by Patrick Macey, Professor of Musicology, in Eastman Schmitt Organ Hall.

Edoardo Bellotti, Associate Professor of Organ, Harpsichord and Improvisation, will give a series of master classes on March 16, 24, 30 and 31. Read more . . .

Nurse visits help at-risk mothers, children live longer, Kitzman's studies show

Over a span of more than 20 years, the research of Harriet Kitzman, the Senior Associate Dean for Research at the School of Nursing, and her colleagues, has proven undisputedly that visits by nurses to disadvantaged new moms and babies lead to many positive outcomes. Not only have the visits been proven to foster healthier pregnancies and deliveries, and improve the health and development of children, but the studies have shown they also help at-risk families improve their economic self-sufficiency.

And last year, the data substantiated yet another dimension of their impact. A study co-authored by Harriet, and published in July 2014's JAMA Pediatrics, revealed that nurse-home visitations actually improve the survival rates of at-risk mothers and their first-born children. Read more . . .

RePORT helps you find NIH-fund research similar to your own

NIH's Research Portfolio Online Reporting Tools, better known as RePORT, provides easy access to info on NIH funded research. The new Matchmaker tool allows you to enter manuscript abstracts, research bios, or other scientific text, and retrieve a list of similar projects from the RePORTER database.

After you submit your text (up to 15,000 characters in length), Matchmaker will analyze it for key terms and concepts, then pull up the top 100 most-similar NIH-funded projects, ranked by match score. Click here to read more.

ORCID distinguishes you from every other researcher

Do you worry about getting credit for your research because your name is common or you have publications under multiple aliases? Do you struggle to keep track of all of your research outputs? Are you annoyed by having to enter the same information over and over in manuscript and grant submission systems? Get an ORCID!

ORCID stands for Open Researcher and Contributor Identification. ORCID is an open, non-profit, global, community-driven initiative to disambiguate authors of scholarly work and to better connect researchers to their research output and to each other. ORCID provides a persistent digital identifier that distinguishes you from every other researcher, and linkages that ensure your work is recognized.

ORCID is increasingly integrated into research systems, including ScienCV. Some funders (such as the Department of Energy) and publishers (such as PLOS and Nature), want you to provide your ORCID id when you submit a proposal or manuscript.

Questions? Contact

Introducing a new faculty member

Ralf Haefner has joined the Department of Brain and Cognitive Sciences as an assistant professor. His primary scientific interest lies in understanding how the brain forms percepts and how it uses them to make decisions, especially in the visual domain. In particular, he is interested in how the brain's perceptual beliefs about the outside world are represented by the responses of populations of cortical neurons, and how their spiking activity gives rise to precepts and decisions. To that end he uses tools from machine learning to construct mathematical models that aim to explain neural responses and behavior. He received a PhD from Oxford University in theoretical astrophysics, an MPA from Harvard University, and worked in the private sector as a management consultant before transitioning to neuroscience.

Congratulations to . . .

Danielle Benoit, Assistant Professor of Biomedical Engineering and of Chemical Engineering, who has received an NSF Faculty Early Career Development (CAREER) award for her proposal "Polymer therapeutics for bone regeneration: next-generation osteoporosis treatments." Benoit's research seeks to develop drug delivery approaches to efficiently and specifically target anabolic drugs to bone to develop novel treatments for osteoporosis. Successful completion of this research will significantly advance therapeutic strategies for osteoporosis, and the approaches developed will be readily adaptable to treat other bone diseases. Read more . . .

University research in the news

Ying Xue, Associate Professor of Nursing, will examine how the nurse practitioner (NP) workforce is spread across the nation — especially in relation to underserved populations — and whether that may be affected by state scope-of-practice regulations. Her project, the "Impact of State Scope-of-Practice Regulation on the Availability of Nurse Practitioners in Caring for Vulnerable Populations," will be funded with a $300,000 grant by the National Council of State Boards of Nursing (NCSBN), a not-for-profit vehicle through which boards of nursing from all 50 states work to promote regulatory excellence for public health, safety and welfare. Read more . . .

For the first time, researchers from the University's Aab Cardiovascular Research Institute (CVRI) at the University of Rochester have used a new gene editing technology called CRISPR-Cas9 to change a regulatory element — a non-protein coding snippet of DNA that controls a gene's expression — in the laboratory mouse. "To our surprise, we found that by making a very subtle change in DNA using the CRISPR-Cas9 technology we were able to virtually wipe out the expression of a gene," said lead study author Joseph Miano, Associate Professor of Medicine and Associate Director of the CVRI. "With this revolutionary technology scientists can literally edit any nucleotide in the 3 billion-plus nucleotides that make up the blueprint of mammalian life forms, including humans." He credits the University's Gene Targeting and Transgenic Facility, led by Lin Gan, Professor of Ophthalmology, for making the research possible. CRISPR-Cas9 is based on a system that bacteria use to defend themselves and their descendants against viruses. Over the last few years, researchers have used this system to precisely alter protein coding genes (genes that provide instructions for making proteins) in human cells, mice, rats, zebrafish, fruit flies and plants. In some cases, disease causing mutations in a protein have been corrected with CRISPR-Cas9, offering some hope for therapeutic intervention. Read more at the Research@URMC blog.

A University study led by Irfan Rahman, Professor of Environmental Medicine, suggests that e-cigarettes are likely a toxic replacement for tobacco products. Emissions from e-cigarette aerosols and flavorings damage lung cells by creating harmful free radicals and inflammation in lung tissue, according to the study, which was published in the journal PLOS ONE and adds to a growing body of scientific data that points to dangers of e-cigarettes and vaping. The investigation suggests the harm begins when the e-cigarette's heating element is activated. The heating element is designed to turn a liquid solution (known as an e-liquid or "juice") into an aerosol that mimics cigarette smoke. The inhaled vapors contain heavy metals and other possible carcinogens in the form of nanoparticles — tiny particulate matter that can reach farther into lung tissue, cell systems, and blood stream. Rahman's study also shows that some flavored e-juices (particularly cinnamon) create more stress and toxicity on lung tissue. Read more . . .

Mark your calendar

Feb. 16: Center for the Integration of Research, Teaching and Learning (CIRTL) Teaching-as-Research poster session. 2 to 4 p.m., Munnerlyn Atrium, Goergen Hall.

Feb. 16: Deadline for AS&E faculty and investigators to file annual reports of outside compensated activity, as required by the University of Rochester Faculty Policy on Conflict of Commitment and Interest and their College. A web-based reporting system supported by the College can be found at

Feb. 17: Exempt Research. Presented by Kathleen Buckwell, Senior Specialist, Office for Human Subject Protection (OHSP). Noon to 1 p.m., Helen Wood Hall Auditorium (1w-304).

Feb. 19: "Mercury Metabolism and Elimination Status in Humans (MerMES): New methods for individualized determination of metabolism of mercury from fish meals," presented by Matthew Rand, Assistant Professor of Environmental Medicine. EHSC Seminar Series, 11 a.m., EHSC Conference Room (Med. Ctr. 4-8820).

Feb. 20: Applications due from new investigators for pilot project funding from the University's Core Center for Musculoskeletal Biology and Medicine. Click here for the full RFA.

Feb. 25: "What's New in Pediatric Bone Health: Recommendations and Resources for the Evaluation of Suspected Bone Disorders in Children," presented by David Weber, Assistant Professor of Pediatrics, Division of Endocrinology and Diabetes. Pediatric Grand Rounds, 8 a.m., Whipple Auditorium (2-6424).

Feb. 25: "Clinical Trials of Devices and FDA Approvals," presented by Arthur Moss, Professor of Medicine. Good Advice: Case Studies in Clinical Research, Regulation, and the Law. Noon, Helen Wood Hall Auditorium (1w-304).

March 1: Deadline for most faculty and other investigators to file annual reports of outside compensated activity, as required by the University of Rochester Faculty Policy on Conflict of Commitment and Interest and their School/College. Eastman School, School of Medicine and Dentistry, and School of Nursing faculty and investigators should use a web-based reporting system supported by their School/College. Links can be found at Laboratory for Laser Energetics, and Warner and Simon school faculty and investigators should use this form. Questions? Contact Gunta Liders or your School/College administrator.

Please send suggestions and comments to Bob Marcotte. You can see back issues of Research Connections, an index of people and departments linked to those issues, and a chronological listing of PhD dissertation defenses since April 2014, by discipline.

Copyright 2013, All rights reserved.
Rochester Connections is a weekly e-newsletter for all faculty, scientists, post docs and graduate students engaged in research at the University of Rochester. You are receiving this e-newsletter because you are a member of the Rochester community with an interest in research topics.