Samples of Student Work 

Student abstracts reproduced with permission. 

Brian Singer, Class of 2016

"To Brux or not to Brux: The Development of Two Novel, Non-Invasive Devices for the Detection of Bruxism"

Bruxism is a disorder in which a patient excessively grinds or clenches their teeth. Symptoms include tooth wear, headaches, back pain, and neck pain. The most common method of treating bruxism is through the use of a mouthguard. The mouthguard does not cure bruxism but only prevents the symptom of tooth wear. Researchers have attempted to reduce bruxism through biofeedback systems. Current bruxism biofeedback devices such as intra-oral pressure sensors and EMG-based systems are intrusive to wear. This study proposes two separate, novel devices that detect bruxism in a less-intrusive manner. The first device is EEG-based and collects data from the F7 electrode located above the left ear. The device uses a machine-learning discriminant-analysis algorithm to detect bruxism from the EEG data. The second device uses Eulerian Video Magnification to amplify temporal color changes in the masseter muscle as seen in a video recording (or live video feed) of bruxism. Both techniques appear to be novel approaches for the detection of bruxism. Both devices were compared to a commercial bruxism detection device to gauge effectiveness and obtain qualitative feedback. Both of the proposed devices demonstrated statistically significant improved efficacy while being less intrusive when compared to the commercially available device.

Jessica Occiogrosso, Class of 2014

"The Involvement of Inflammation in the Intervertebral Disc’s Altered Response to Mechanical Load"

Degenerative disc disease (DDD) is a condition in which a patient experiences pain from a damaged intervertebral disc. This disease affects nearly 80% of the population (Kelly, 2012). Surgical treatments for DDD focus on immediate symptom alleviation as opposed to restoration of the disc’s anti-catabolic phenotype. With this study, cow tail disc tissue sections were stained using antibodies specific for two different proteins – an integrin and an ion channel (two different types of molecules that allow the disc to come in contact with its external environment). The expression of these proteins between groups where the discs were exposed to different stimuli- a pro-inflammatory cytokine and/or a mechanical load- were compared in order to ultimately draw conclusions about their roles in the onset of disc degeneration. The results showed that, when comparing the percentage of the inflammatory molecule (TNF?) that got into the cells with the subsequent percentage expression of either protein, there was a statistically significant positive correlation between the expression of TNF? and increased expression of TRPV4. This suggests that TRPV4 may play an important role in the pathogenesis of disc degeneration, suggesting that this molecule could be targeted to stop pain associated with DDD. 


Mentors: Dr. James Iattridis and Benjamin Walter

Mt. Sinai Medical School


Adam Ingber, Class of 2014

"Cerebrospinal fluid biomarkers and reserve variables as predictors of future “non-cognitive” outcomes of Alzheimer’s disease"


Alzheimer’s disease (AD) is a devastating neurodegenerative disease that affects a steadily increasing portion of the elderly. It is imperative that early detection and treatment strategies be developed to identify the disease in its early stages and begin potential therapeutic options to halt or prevent conversion from preclinical to symptomatic AD. Using longitudinal data from participants enrolled in studies at the Washington University Knight ADRC, I used linear mixed models to examine the way in which cognitive and brain reserve variables mediate how AD biomarker levels in cognitively normal persons predict future changes in function, weight, mood, and behavior. While education was not shown to have a significant effect on predicting future non-cognitive decline with time, total brain volume exhibited a strong and significant effect when combined with biomarker values to predict decline due to AD over time. My findings suggest that brain reserve plays a stronger role than cognitive reserve in building protection against non-cognitive impairment in AD. This study will contribute to the growing literature on predictive biomarker models of AD and may aid in the future development of individualized risk profiles that can predict future consequences of AD years in advance, with tremendous potential medical and financial implications.


Mentor: Dr. Catherine M. Roe

Washington University School of Medicine

Nicole Pollack, Class of 2011

"Determining Optimal Characteristics of Monoclonal and Polyclonal Anti-METH Antibodies Through Affinity Analysis"

In immunotherapy, if an addict self-administers METH the antibodies will bind with high affinity to the METH molecules and prevent them from entering the brain; this eliminates METH’s euphoric effects. Without reward, the addict can more quickly learn not to use METH. Antibody affinity is measured by the magnitude of the dissociation constant (KD) of antibody in the presence of METH, with a lower KD indicating higher affinity. Based on a theoretical model, it was hypothesized that the most effective anti-METH antibodies would have KD values < 10 nM. However, antibodies for passive immunity tested under simulated human physiological conditions (37°C; pH = 7.3; 0.15 mM NaCl) showed KD values as much as five-fold higher than when KD values were determined at 4°C (standard laboratory condition). Thus, antibodies must have a very low initial KD value (< 2 nM) so that despite the KD increase due to temperature and pH change, the antibody will still be able to confine METH to the serum based upon the theoretical modeling. In this study, KD values were obtained for antibodies generated from active immunization using different antigen synthesis conditions. I found that synthesis conditions with a maleimide to hapten ratio of 240:50 were optimum. Only one of the antibodies in this study had a low enough KD value based on the model to fully contain METH in the serum. Future research should focus on optimizing chemical synthesis of the antigen to generate antibodies tolerant of temperature and pH changes with KD values < 10 nM. 

Mentor: Dr. Michael S. Owens
University of Arkansas for Medical Sciences

Connor Berlin, Class of 2011

"Examining the Influence of Constitutively Activated Akt on Schwann Cell Myelination"

In the most common demyelinating disorder, Multiple Sclerosis (MS), oligodendrocytes of the CNS are damaged and killed by the host's own immune system. While current treatments such as beta interferons have demonstrated an ability to slow progression of MS, the damages to CNS myelin (and concurrent loss of axon function) still remain. The goal of future cellular therapies is to remyelinate CNS axons in which myelin has been completely lost. Schwann cell precursors are good candidates for these cellular therapies. However, the degree of CNS myelin repair achieved by transplanting Schwann cell precursors is still limited. Thus, it has become essential to understand the role of specific molecules/signaling pathways that promote Schwann cell differentiation, survival, and myelination in vivo. This paper examines the potential influence of a specific protein kinase in Schwann cell myelination, known as Akt. To determine Akt's influence, sciatic nerve samples from CNP-MyrAkt and wild type mice were extracted. I teased apart the individual fibers, stained the nerves, and measured for myelin sheath lengths and diameters. Significance of change was analyzed using t tests and ANOVAs. The results revealed that transgenic myelin sheath length was significantly shorter. The data therefore suggested that Akt was acting as a pro-survival factor for Schwann cells.

Mentor: Dr. James Salzer
Salzer Lab, NYU

Lily Kosminsky, Class of 2010

"Assessing the Risk of Stress Shielding in Osseointegrated Transfemoral Implants"

Lower limb prosthetics are currently being developed that attach directly to the skeleton via osseointegration, the phenomenon whereby bone bonds to titanium. This is a promising alternative to traditional socket attachment, which is not an option for patients with high level amputations. One potential problem with osseointegrated prostheses is stress shielding, whereby an implant removes load from the bone, thus weakening it. Stress shielding is known to cause long-term problems including osteonecrosis and aseptic loosening in osseointegrated joint replacements but has not yet been studied in osseointegrated femoral implants. This study identified stress shielding resulting from femoral implants in an ovine model. Both bone porosity and mineral apposition rate, accepted gauges of stress shielding, were analyzed for each implanted bone and compared to the control group. Porosity and mineral apposition rate were greater in the implanted bone, indicating stress shielding did occur. Stress shielding has already been recognized and addressed within the field of osseointegrated joint replacements. This novel study indicates that stress shielding should be considered during the further development of osseointegrated lower limb prosthetics and that the bone quality of patients who are already using early versions of these prosthetics should be closely monitored.

Mentor(s): Dr. Roy D. Bloebaum
University of Utah Medical Center

Rachel Cawkwell, Class of 2010

"Effect of Tumor Microvesicles on Macrophages in Cancer"

A normal, though understudied, physiological process is the shedding of microvesicles, membranous sacs, from cells. These microvesicles are representative of the cell type they are derived from and can transfer membrane receptors, proteins, mRNA, and organelles in the derived cell. Microvesicles are found in higher numbers in cancer patients and appear to be a vital method of communication for tumors. This project studies how microvesicles work as tumor messengers with regard to macrophages. Macrophages are a type of white blood cell that the tumor uses to direct processes such as blood vessel recruitment and invasion of other tissues. Tumor microvesicles were isolated to see whether they could transfer mRNA to macrophages and increase macrophage proliferation and migration. The tumor microvesicles did transfer mRNA to macrophages as confocal microscopy and qRT-PCR revealed uptake of fluorescence expressing mRNA from microvesicles. A proliferation assay showed no significant change in macrophage proliferation, but an invasive assay demonstrated that tumor microvesicles can increase macrophage migration. This means that tumor microvesicles could potentially recruit macrophages to the primary tumor site, perhaps by transferring mRNA. Further investigation into the details of tumor microvesicle-macrophage communication should confirm and extend these results.

Mentors: Dr. David Lyden and Dr. Hector Peinado Selgas, Weill Medical College of Cornell

Samantha Sanders, Class of 2009

“Efficient Silencing of the Bcl-2 Oncogene through the Unassisted Delivery of Antisense Oligodeoxyribonucleotides G3139 and SPC2996”

Oligodeoxyribonucleotides, or short molecules comprised of DNA sequences, can effectively silence specific genes by inhibiting mRNA translation. Two common oligodeoxyribonucleotides are G3139 and SPC2996, both of which target the Bcl-2 oncogene.  These oligos had previously been delivered with a lipid carrier to offset hypothesized interference posed by the cell membrane. We attempted to disprove the necessity of such a carrier by evaluating the efficacy of unassisted delivery. When delivered without lipid carriers, G3139 and SPC2996 produced Bcl-2 protein and mRNA downregulation in multiple cell lines. These findings indicate that unassisted delivery facilitates oligonucleotide gene silencing. G3139 and SPC2996 also produced few off-target effects, i.e. no downregulation in non-targeted proteins.  Further investigation concerning naked oligonucleotide uptake in cells also revealed that oligonucleotides work with the cytoplasmic enzyme Ago2 to inhibit mRNA translation, though the RNase H enzyme had been thought previously to play this role. 

Mentor:  Cy A. Stein, M.D., Ph.D. Professor of Medicine, Urology, and Molecular Pharmacology    Montefiore Medical Center
University Hospital of the Albert Einstein College of Medicine

Rachel Schuman, Class of 2009

"Identifying Antigen Markers on the Cell Surface of Cancer Stem Cell Population Isolated from Ovarian Cancer Cell Lines OVCAR3 and OVCAR5"

Cancer stem cells (CSCs) found in multiple types of cancer, are a small population of cells that are hypothesized to be responsible for initiation, maintenance and metastasis of cancer. It has been demonstrated that conventional cancer therapy, specifically chemotherapy, fails to completely eradicate the CSC population. Two populations, CSCs and resulting differentiated cells, were isolated from ovarian cancer cell lines OVCAR3 and OVCAR5 in this experiment. These experimental cell lines were exposed to known primary antibodies, and analyzed using flow cytometry (FACS). The antigen-antibody complexes that formed revealed specific protein markers present on the cell surface of both populations. GM2 and Muc1 were of interest as clear differences in expression were observed when comparing the two populations of cancer cells with respective to OVCAR3 and OVCAR5.  Results showed that as CSCs differentiate, their molecular nature changes; either antigen expression is maintained, a new antigen reveals itself, or antigen expression is lost. This demonstrates that there is a clear distinction between the two populations. The identification of GM2 specifically on the CSC surface, contributes towards the development of a more advanced cancer vaccine that will target both the CSC population and the differentiated cell populations in order to stimulate a stronger immune response.

Mentor(s):  Dr. Philip Livingston and Mr. Dean George, 
Memorial Sloan-Kettering Cancer Center

Stephanie Doctor, Class of 2009 

"Handedness in Grooming Behavior of Gelada Baboons (Theropithecus gelada) and its Relation to the Emergence of Language in Humans"

Allo-grooming, or the grooming of others, is a mechanism for creating and servicing relationships, which in turn create alliances that are essential for survival. These alliances allow primates to form larger groups, which are beneficial in defense against predators. Groups face a cognitive limit on the number of relationships any individual can keep track of. However, the efficiency of grooming restricts group size further – at a point the groups become so large that maintaining them becomes a time burden, and the groups may be forced to divide. Therefore, grooming by itself is not efficient enough to allow groups to reach full cognitive potential. Dunbar (1993) proposed that language emerged in early humans as a more efficient mechanism for bonding, especially in the context of larger social networks. As language and handedness are both somewhat lateralized functions in the brain, similar hemispheric control of the two would suggest a common origin, and therefore would be tempting support of Dunbar’s proposal.  This observational study, which I created independently, investigates handedness in allo-grooming of gelada baboons, in search of a common origin in brain control with language. Gelada behavior was recorded on camera at the Bronx Zoo. The data suggest a preference for right-handedness. This pilot study should be replicated with a greater subject pool to allow for in-depth quantitative analysis of this trend.

Mentor:  Florence Klecha, LVT
Bronx Zoo, Bronx, NY

Eli Kosminsky, Class of 2009

"World of Warcraft: The Viability of Massively Multiplayer Online Role-Playing Games as Platforms for Modeling and Evaluating Perfect Competition"

World if Warcraft (WOW) is the most popular online game in history, with over 11 million players across the world.  The interactions of these players have formed a complex, virtual economy which in many ways behaves like a real world economy.  The objective of this study was to determine whether the economy of WOW is a competitive free maket approaching the ideal of perfect competition, and if so, whether it constitutes a suitible platform for further economic research.  Data was collected detailing the prices at which selected virtual goods were sold in WOW auction houses, over a period of thirty days.  The research concluded that the virtual economy of WOW in most respects behaves like a highly compeitive real world market, and in fact approaches the ideal of perfect competition.


Max Dichter, Class of 2009

"Aerodynamic Efficiency of a Formula Car Front Wing"

Aerodynamics is the study of the motion of fluid around an object by which forces are created. Aerodynamics is crucial to Formula One (F1) car racing as it plays a very influential role in the design of the F1 front wing. In this study, Max tested the aerodynamics of a Formula Renault 98’ front wing. Prior to the study, scale model wing profiles had been constructed out of glass nylon. These were replicated in this study using a Computer Aided Design (CAD) program called SolidWorks.  Wing aerodynamics was then tested in a wind tunnel and compared to results from a Computational Fluid Dynamics program (FloWorks).  Future research will translate the best design into a full-scale mockup.

Mentor: Dr. Craig Dawson, Automotive Engineering Design
Oxford Brooks University, Oxford, U.K.

Kristy Gardner, Class of 2008

"Nutrient Limitation and Autotoxicity in the Red Tide Dinoflagellate Alexandrium fundyense"

The dinoflagellate Alexandrium fundyense causes harmful algal blooms in the Gulf of Maine that have devastating human health and economic impacts.  This study addressed the effects of nutrient deficiency and autotoxicity in A. fundyense.  A. fundyense was grown in nine treatments with variable concentrations of nutrients (f/2, f/4, f/20, based on Guillard’s f/2 seawater enrichment) and spent media (0, 50%, 100%) to form a 3x3 factorial system.  Data from cell counts and the intracellular probes SYTOX, ELF-97, and SNARF taken over a six week period revealed that the initial division rate was suppressed by percentage spent media in the f/2 and f/4 treatments, but by nutrient deficiency in the f/20 treatments; nutrient limitation was dominant during a second exponential growth phase.  Depressed photosynthesis activity and high nitrite concentrations in spent-media treatments suggest that autotoxicity may interfere with nitrite reductase activity, causing nitrogen limitation.  Nitrogen limitation induces sexual reproduction in A. fundyense, which causes cyst formation, allowing the cells to survive the winter.  This study therefore suggests that autoinhibition in A. fundyense is a self-regulatory measure to allow for survival.

Mentor: Dr. Gary Wickfors
Northeast Fisheries Science Center

Jonathan Kaufman, Class of 2008

"Examining STR-based Genetic Deviation between Indigenous East Asian Populations and their Immigrant Counterparts to Determine the Credibility of Immigrant Derivation in Genetic Anthropology"

To have the potential to examine the intricacies of human history via tangible genetic evidence (through the statistical analysis of genetic samples) has permitted scholars to gain a renewed perspective of previously-unknown historical phenomena. Globalization and urbanization have created a tapestry in which previously-rural native populations have been largely (and in some cases, completely) absorbed into foreign cities. The validity of using data from these immigrants in genetic anthropology studies is questioned as a result of potential founder effects and genetic drift. This research demonstrates that Short Tandem Repeat (STR) polymorphism frequency data extracted from urban second and third generation East Asian immigrant populations (of Japanese, Chinese, Korean, and Vietnamese origin) worldwide generally expresses the same average genetic composition as data retrieved from their indigenous counterparts (from the native nation). Genetic anthropology is thus viable in a situation of overwhelming immigration and urbanization. Genetic frequency data was attained from Yale University’s ALFRED (Allele Frequency Database), a global effort to consolidate forensic and anthropological genetic data onto a single interface.

Mentor: Dr. Kenneth Kidd 
Yale University School of Medicine

John Granata, Class of 2007

"Improving the Information Transfer Rate of the P300 Speller Brain Computer Interface"

The primary goal of brain-computer interface research is to provide new communication and control alternatives for people with severe neuromuscular disorders that cannot communicate with the external world through normal muscular output channels. The Wadsworth Center P300 Speller brain-computer interface (BCI) enables a user to spell words on a computer screen by detecting patterns in neuronal activity in response to characters flashed on-screen. The classification problem involves identifying which character, in a matrix of flashing on-screen characters, a user is focusing attention on. By detecting which character elicits a P300 brain rhythm, an evoked electrical potential time-locked to the intensification of the character being attended to, an investigator can isolate upon which character in the on-screen matrix a user is focusing attention on, and print this letter to the screen. 

In order to detect the presence of a P300 rhythm, the goal is to identify a subset of spacio-time points, extracted from the recorded brain responses to the flashing on-screen characters, that best discriminate between the characters that generate P300 rhythms and those that don’t.   In order to address this classification problem, I designed, programmed, and tested new software for the P300 Speller BCI to select the features (spacio-time points) that best predict presence of a P300, and then classify the brain responses as efficiently as I could. All implementation was done in the Matlab 7.0 programming language. After extracting the brain responses associated with the flashing of each character and applying standard filtering and preprocessing to this data, I applied a stepwise removal algorithm to the responses in order to select the “m” most significant time-points for predicting P300 presence. I then implemented a novel leaps and bounds feature selection algorithm, which I applied to these “m” variables, in order to generate the “m”-best subsets of time-points that maximized the Bhattacharya distance between target (P300) and non-target (non-P300) data. A linear multiple regression model was then trained for each of the selected subsets of time-points (features) using class labels {1,-1} for targets and non-targets, respectively. These regression models were then used to predict which character in the matrix of characters that the user was focusing attention on, and results from each model were cross-validated to ensure accuracy. Accuracy was compared to the current system in use at the Wadsworth Center in Albany. To address the issue of increasing speed of the P300 Speller brain-computer interface, a new soft flashing boundary was also implemented in order to make more confident character predictions after obtaining the minimum amount of information necessary from the user. These objectives addressed improving the speed and accuracy of the Wadsworth Center P300 Speller BCI, and the results show that both the new feature selection algorithms and the new intensification boundary increase the information transfer rate (correct characters communicated per minute) of the BCI. Furthermore, the novel leaps and bounds feature selection algorithm shows great promise for the future, and I am currently developing the algorithm to be able to handle more predictor variables than currently shown in this study. 

Mentor: Dr. Theresa Vaughan
Wadsworth Center, NYSDOH, Albany, NY

Byram Hills Central School District | 10 Tripp Lane, Armonk, NY 10504 | (914) 273-4082