Mathematics
Calculating a Critical Distance for Metastasis
Sam Antill ('13); Mentor: Ami Radunskaya
Abstract: In this work we seek conditions under
which a tumor’s growth will be contained and
metastasis will not occur. We hypothesize that
there exists a critical distance from the blood
vessel for which there will be insufficient nutrients
for metastasis. In this model, we focus on the
qualities of tumor gluttony and adhesion rather
than tumor immune interactions as a means of
potentially impeding tumor growth. We develop a
hybrid cellular automata model to simulate tumor
growth and using this model we find that a critical
distance from the initial tumor to the nearest blood
vessel does exist. Our results show that tumors
that begin further than this distance from a vessel
will not metastasize. The results indicate a critical
distance for the model that agrees with
experimentally determined data. We conclude that
distance between a tumor and the nearest blood
vessel can be an important factor in tumor
prognosis.
Funding Provided by: The Norris Foundation
Microarray Data Analysis: Finding Life’s Music through Statistical Noise
Joseph Replogle ('13); Mentors: Johanna Hardin, Laura Hoopes
Abstract: Microarray technology measures thousands
of genes’ transcription levels
simultaneously. Analyzed properly, microarray
data allows biologists to identify differential
expression in cells from scientifically interesting
experimental conditions. Due to the heterogeneity
of microarray experimental designs, scientists
choose from numerous analysis techniques to
obtain proper results. Therefore, in my SURP I
sought to integrate normalization, differential
expression identification, and gene ontology (GO)
techniques to analyze yeast microarray data
produced in summer 2010 by Hoopes’ lab. Using
limma and GOstats R packages, I built off the
methods of Yui et al. including loess scaling for
normalization and an empirical Bayes method to
identify differential expression. Additionally, I
included GO terms described in Falcon et. al. I
analyzed data from 10 Saccharomyces cerevisiae
microarrays to identify differential expression and
GO terms. The methodology used this summer
could be applied to future experiments in
identifying differential ex-pression and GO terms
in yeast or other organisms.
Funding Provided by: The Fletcher Jones Foundation