Welcome to emmahodcroft.com
I am a post-doctoral researcher at the University of Edinburgh in Scotland, working with Prof Andrew Leigh Brown. I recently completed my PhD, during which I developed a new, quantitative-genetics based approach to estimate the heritability (viral genetic effect) of virulence in HIV.
I am now primarily involved in the PANGEA_HIV
initiative, funded by the Bill & Melinda Gates Foundation. PANGEA_HIV aims to use
phylogenetics and molecular epidemiology to gain new understanding of the HIV epidemic
in sub-Saharan Africa. (More information in 'Research' below.)
The most exciting phrase to hear in science, the one that heralds
the most discoveries, is not "Eureka!" but "That's funny..."
Current
My research currently focuses on HIV. I am part of the PANGEA_HIV
initiative, funded by the Bill & Melinda Gates Foundation. PANGEA_HIV aims to use
phylogenetics and molecular epidemiology to gain new understanding of the HIV epidemic
in sub-Saharan Africa.
I have created a stochastic, agent-based model (DSPS-HIV) that simulates HIV epidemics, which I've used to generate data sets that can be used to assess phylogenetic methods. The DSPS-HIV is based on Samantha Lycett's
Discrete Spatial Phylo Simulator code,
but has been highly modified to enable it to simulate realistic HIV epidemics. Disease stage and transmission
risk are dependant on viral load, and contact networks are highly customizable. Code for the DSPS-HIV will be
available online in the future.
You can see the simulated datasets released as part of PANGEA_HIV work package 4
here. You can read more details about the DSPS-HIV,
as well as a full description of the PANGEA_HIV methods comparison exercise in our
recent publication in MBE.
I am currently working on modifying the DSPS-HIV to better simulate more complex HIV epidemics - specifically Western heterosexual and men-who-have-sex-with-men (MSM) epidemics - in order to investigate questions such as how contact networks and sampling bias may influence phylogenetic analysis.
I recently completed my PhD, during which I developed a new method for estimating the heritability of viral load in HIV. Though the influences of many host and environmental factors on viral load are well understood, the role of the viral genome itself in determining viral load is less clear.
I adapted a well-established method from population genetics to more accurately estimate the heritability of viral load using a phylogeny of viral sequences. This method enables analysis on incredibly large datasets, and I have investigated the viral genetic contribution to viral load in subtypes B and C in the UK HIV epidemic, using 8,483 and 1,821 sequences, respectively (provided by the UK HIV DRB). Papers and presentations resulting from this work can be found in 'Publications and Presentations' below.
"But all evolutionary biologists know that variation itself is nature's only irreducible essence. Variation is the hard reality, not a set of imperfect measures for a central tendency. Means and medians are the abstractions."
Post-Graduate
-
For my Master's (MSc) dissertation, I looked for evidence of adaptive selection in coding and
non-coding genes in Drosophila. Adaptive substitution rates in coding regions and
5' and 3' UTRs (untranslated regions) were analysed by tissue-specific, time-specific,
and immune-related gene function.
Coding regions of immune-related genes were found to have significantly higher adaptive rates than non-immune-related genes, but no difference was found in UTRs. All three regions were shown to have similar rates of adaptive evolution in most tissue-specific and time-specific genes, though UTRs had significantly higher adaptive rates than coding regions in some cases. The study provided evidence that UTRs have a faster overall adaptive rate but also more non-adaptive substitutions, and that the adaptive rate of UTRs and coding regions varies by gene function.
Graduate/Under-Graduate
-
After graduating from TCU, I worked with Dr. John Horner on the carnivorous pitcher plant
Sarracenia alata. I aided in preliminary studies on the aromatic compounds that
Sarracenia may use to attract prey, and also investigated the genetic variation
in four geographically separate populations of Sarracenia. Using AFLP analysis,
our study concluded that though long suspected to be primarily clonally reproducing,
only 14% of the genetic variation in Sarracenia alata occurred among populations,
while 86% occurred within populations, indicating that clonal spread is actually quite low
in these populations.
I presented a poster on this work at the Evolution 2010 Conference in Portland, Oregon.
"My work can be taxing and hazardous, but dull? Never."
Publications
Phylogenetic Tools for Generalized HIV-1 Epidemics: Findings from the PANGEA-HIV Methods Comparison. Ratmann O, Hodcroft EB, et al.,
on behalf of the PANGEA-HIV consortium. Molecular Biology and Evolution. 2016.
(link)
Two models were used to create a variety of HIV epidemic simulations. Sequences and phylogenies were publicly released and
groups were invited to try and estimate epidemic parameters such as incidence, transmissions during acute stage, and migration rate.
The Contribution of Viral Genotype to Plasma Viral Set-Point in HIV Infection. Hodcroft EB, Hadfield JD, Fearnhill E,
Phillips A, Dunn D, O'Shea S, Pillay D, Leigh Brown AJ. PLoS Pathogens. 2014.
(link)
Here we implement a new phylogenetic method to estimate the heritability of viral load in subtype B in the UK, and investigate
the change in viral load over time due to selection.
Automated Analysis of Phylogenetic Clusters. Ragonnet-Cronin M, Hodcroft EB, Hue S, Fearnhill E, Delpech V,
Leigh Brown AJ, Lycett S, on behalf of the UK HIV RDB. BMC Bioinformatics. 2013.
(link)
Two new programs are introduced to allow efficient and easy analysis of phylogenetic clusters. The
ClusterPicker allows
users to 'pick' clusters of closely related sequence by specified thresholds; the
ClusterMatcher (written by myself) allows
users to 'match' clusters containing the same sequences from two different runs, and also to investigate the attributes of
the clusters.
Presentations
3 Minute Thesis
In 2014 I had the privilege of competing in the 3 Minute Thesis competition,
presenting my work on estimating the heritability of
viral load in HIV. I advanced through the School and College levels to win both first prize and the 'people's choice' award at the
University of Edinburgh finals. I then went onto the UK Semi-Final in York, where I advanced to the UK final in Manchester alongside
five others. I also prepared a video of my presentation to compete against 17 other finalists in the world-wide
Universitas 21 competition,
where I placed 3rd.
You can view a video of my 3 Minute Thesis below!
Posters and Talks
Detecting Changes in Incidence Using Phylogenetic Tools: Simulation-Based Studies Within the PANGEA_HIV Initiative. Presented at CROI 2015 in Seattle, WA (abstract and poster available here).HIV Virulence Has Not Increased in the UK Subtype B Epidemic. Presented at CROI 2014 in Boston, MA (abstract and poster available here).
I have also presented my work on heritability estimation in subtypes B and C at CROI 2012 in Seattle, WA, and CROI 2013 in Atlanta, GA. Unfortunately these posters are no longer available online. I've also spoken about this work at many conferences, including the 45th Population Genetics Group Conference in Nottingham, England, and the 19th, 20th, and 21st HIV Dynamics and Evolution Conferences in Asheville, NC; Utrecht, the Netherlands; and Tucson, AZ.
My love of programming means I'm always eager to code something. In the course of my research,
I've written a few programs to aid me in my tasks.
In particular, TreeCollapserCL, which collapses trees based on bootstrap support values, is the most popular program
and potentially the most useful to others.
All of these programs are Java, but I very much like R, and have dabbled in Perl and Python.
Recent
*Most Popular* TreeCollapserCL 4
A new, improved version of TreeCollapseCL that can root trees and find lengths of branches
and average bootstraps of nodes, as well as collapsing nodes with bootstraps
below a user-specified threshold.
Updated: Corrects an issue with the collapsing algorithm that sometimes lead to over-collapsing. It's
highly recommended that you re-run data with TreeCollapseCL4.
PareTree 1.0.2
A basic, command-line Java program that allows users to 'pare' down their tree by either
removing unwanted sequences/leaf-nodes, removing bootstrap information, removing branch lengths - or any combination
of those three - quickly and efficiently.
Updated: Now also allows users to remove branch lengths from the tree. Also fixed a few minor bugs.
*Published* ClustMatcher
A cluster is a monophyletic group of sequences in a phylogeny that fall within specified
bootstrap support and genetic distance thresholds. In the study of infectious diseases,
especially HIV, they can represent transmission events between individuals.
Samantha Lycett's tool, ClusterPicker, is able to 'pick' clusters from a phylogeny.
The ClustMatcher tool can then be used to find clusters the contain some or all of the same
sequences between the two data sets, and outputs annotated FigTree files containing matching
clusters. This allows the change in cluster size to be compared over time.
ClustMatcher can also be used with one data set to select only clusters that contain a certain
number of sequences or have a certain attribute (clusters that contain females, for example), for
further study.
The paper detailing the ClusterMatcher and ClusterPicker software is here.
DSPS-HIV
Based on the Discrete Spatial Phylo Simulator, coded by Dr. Samantha Lycett, the DSPS-HIV is a stochastic, agent-based model which has been highly modified to simulate realistic HIV epidemics. Transmission risk and disease progression rate are dependant on viral load, which is heritable, and contact networks are highly customizable. Acute, chronic, and AIDS disease stages are modelled, and treatment can be introduced at varying levels and speeds. All transmissions are tracked, so that a viral phylogeny of the epidemic is produced.
SeqMatcher
A program written to match sequences across HIV database downloads. Because sequence and
patient names are re-anonymised before each download release, it has so far been impossible
to track the differences in how the sequences cluster from download-to-download. This program
attempts to identify matching sequences between downloads. Because sequences are thousands of
characters long, and each download contains tens of thousands of sequences, running efficiency
and memory allocation are important in this program.
Unfortuantely, because this program is written specifically for the HIV database, it would not
be useful for me to provide it here for public use.
TreeCollapserCL 3.2
Outdated! This version has an error in the collapsing algorithm that can sometimes lead to over-collapsing.
I am leaving it up temporarily so that users can compare trees run with the old algorithm to trees run with TreeCollapseCL4.
Please use TreeCollapseCL4 instead. Nothing else was changed in this update, so running this old version will not
solve any other issues.
Older
Unfortunately, Java has changed since I wrote these programs years ago! It no longer allows copy/paste into or out of Java Applets, and I don't have time to re-code these right now - so for the moment they are unusable!
TreeCoder
Allows the user to copy a visual phylogenetic tree (say out of a paper) into phyloXML or Nexus-type code easily so they can manipulate it. Might be potentially useful, though only in limited circumstances, since I doubt many people do this regularly. It was really useful (and originally written) for a paper I wrote on coevolution of hantavirus during my MSc. (I used it to point out that a published tanglegram had been left partially tangled to support the paper's premise, but could actually be resolved much better than the author seemed to be implying.)
Word Counter
Does what it says, but doesn't count references in parenthesis!
I am rarely happier than when spending an entire day programming my computer to perform automatically a task that would otherwise take me a good ten seconds to do by hand.
About Emma Hodcroft
"I cannot unshackle myself from argument [...] I was born into argument. Argument was my first nursemaid. Argument is my lifelong bedfellow. What’s more, I believe in argument and I even love it. Argument is our most steadfast pathway toward truth, for it is the only proven arbalest against superstitious thinking, or lackadaisical axioms."
Academic
I completed my undergraduate degree in biology at Texas Christian University (TCU), where I helped to set up and run the Purple Bike Program, a green initiative that rented free bikes to students to help reduce pollution and carbon emissions on campus. I also worked as a Java programming tutor, a job I very much enjoyed.
After graduating in December of 2008, I took a research assistant position with Dr. John Horner investigating how carnivorous Sarracenia alata pitcher plants attract their prey, as well as the genetic diversity of Sarracenia populations in the Southern US. (Obligatory research-in-action photo to the left.)
In the autumn of 2009, I moved from Texas to Edinburgh, Scotland, and began my master's degree at the University of Edinburgh on the Quantitative Genetics and Genome Analysis course. Though a challenging year, the course gave me an excellent introduction to the world of population and quantitative genetics. During the course, I prepared reports and did research projects on the evolution of SIV & HIV, the co-evolution of hantavirus and its hosts, genes involved in vitamin D production and colon cancer, and, for my final dissertation, the evidence of adaptive selection in coding and non-coding DNA in Drosophila.
After receiving my MSc degree with distinction in the autumn of 2010, I took a year-long research assistant position with Prof. Andrew Leigh Brown investigating virulence in HIV. Having been thoroughly won-over by the wonderful world of viruses, I began my PhD with Prof. Leigh Brown in September 2011 to continue my work on HIV. (More information in the 'Research' section.)
I have begun my post-doc position with the PANGEA_HIV initiative, continuing in Andrew Leigh Brown's lab, where I am developing a realistic, stochastic agent-based model to simulate HIV epidemics in sub-Saharan Africa (again, more information in the 'Research' section.)
Personal
Born in Norway, and raised spending half the year in Scotland with my father and half the year in Texas with my mother, I'm a strange mix of two countries more similar than one might expect!
My half-and-half upbringing has given me a unique perspective on life, as well as an interesting vocabulary and an amusing accent. A fan of both kilts and cowboy boots, I feel equally at home in both places.
I'm lucky enough to have had the opportunity to travel around North and South America, Europe, and even venture a little into Asia. My bi-annual migrations between Texas and Scotland all my life mean I'm quite at home in airports and on planes, and am no stranger to travel at all.
As well as my love of biology and evolution, I'm an armchair sociologist and feminist, and very much enjoy a good debate on any controversial topic. I love reading a wide variety of books, from popular fiction and 'pop-sci' to non-fiction and classics. Being a third-generation computer geek, I enjoy all things tech-y and have had a deep love of programming since 15.
I played violin regularly in various orchestras from age 10 to 21 and still enjoy it, though I don't play as much as I'd like to at the moment. Like everyone else on the planet, I enjoy photography. Finally, I have a fondness for the colour purple, cephalopods, potatoes, and cats.
I can be contacted at the address and phone number below:
Phone: 0131 650 8683 (International: +44 131 650 8683)
The prevalence of spam-bots keeps me from posting my email address, but you can contact me via the feedback form.
“Science may never come up with a better office communication system than the coffee break.”
