Thesis/thesis.lyx

#LyX 2.3 created this file. For more info see http://www.lyx.org/
\lyxformat 544
\begin_document
\begin_header
\save_transient_properties true
\origin unavailable
\textclass extbook
\begin_preamble
\usepackage{fancyhdr}
\pagestyle{fancy}
\renewcommand{\headrulewidth}{0pt}
\rhead{}
\lhead{}
\rfoot{}
\lfoot{}
\cfoot{\thepage}
\usepackage{draftwatermark}
\end_preamble
\use_default_options true
\maintain_unincluded_children false
\language english
\language_package default
\inputencoding auto
\fontencoding global
\font_roman "default" "default"
\font_sans "default" "default"
\font_typewriter "default" "default"
\font_math "auto" "auto"
\font_default_family default
\use_non_tex_fonts false
\font_sc false
\font_osf false
\font_sf_scale 100 100
\font_tt_scale 100 100
\use_microtype false
\use_dash_ligatures true
\graphics default
\default_output_format default
\output_sync 0
\bibtex_command default
\index_command default
\paperfontsize 12
\spacing double
\use_hyperref false
\papersize letterpaper
\use_geometry true
\use_package amsmath 1
\use_package amssymb 1
\use_package cancel 1
\use_package esint 1
\use_package mathdots 1
\use_package mathtools 1
\use_package mhchem 1
\use_package stackrel 1
\use_package stmaryrd 1
\use_package undertilde 1
\cite_engine basic
\cite_engine_type default
\biblio_style plain
\use_bibtopic false
\use_indices false
\paperorientation portrait
\suppress_date false
\justification true
\use_refstyle 1
\use_minted 0
\index Index
\shortcut idx
\color #008000
\end_index
\leftmargin 1.5in
\topmargin 1in
\rightmargin 1in
\bottommargin 1in
\secnumdepth 3
\tocdepth 3
\paragraph_separation indent
\paragraph_indentation default
\is_math_indent 0
\math_numbering_side default
\quotes_style english
\dynamic_quotes 0
\papercolumns 1
\papersides 2
\paperpagestyle default
\tracking_changes false
\output_changes false
\html_math_output 0
\html_css_as_file 0
\html_be_strict false
\end_header

\begin_body

\begin_layout Title
Bioinformatic analysis of complex, high-throughput genomic and epigenomic
 data in the context of immunology and transplant rejection
\end_layout

\begin_layout Author
A thesis presented 
\begin_inset Newline newline
\end_inset

by
\begin_inset Newline newline
\end_inset

 Ryan C.
 Thompson
\begin_inset Newline newline
\end_inset

 to
\begin_inset Newline newline
\end_inset

 The Scripps Research Institute Graduate Program 
\begin_inset Newline newline
\end_inset

in partial fulfillment of the requirements for the degree of
\begin_inset Newline newline
\end_inset

 Doctor of Philosophy in the subject of Biology
\begin_inset Newline newline
\end_inset

 for
\begin_inset Newline newline
\end_inset

 The Scripps Research Institute
\begin_inset Newline newline
\end_inset

 La Jolla, California
\end_layout

\begin_layout Date
May 2019
\end_layout

\begin_layout Standard
[Copyright notice]
\end_layout

\begin_layout Standard
[Thesis acceptance form]
\end_layout

\begin_layout Standard
[Dedication]
\end_layout

\begin_layout Standard
[Acknowledgements]
\end_layout

\begin_layout Standard
[TOC]
\end_layout

\begin_layout Standard
[List of Tables]
\end_layout

\begin_layout Standard
[List of Figures]
\end_layout

\begin_layout Standard
[List of Abbreviations]
\end_layout

\begin_layout Standard
[Abstract]
\end_layout

\begin_layout Chapter*
Abstract
\end_layout

\begin_layout Chapter*
Introduction
\end_layout

\begin_layout Section*
Background & Significance
\end_layout

\begin_layout Subsection*
Biological motivation
\end_layout

\begin_layout Itemize
Rejection is the major long-term threat to organ and tissue grafts
\end_layout

\begin_deeper
\begin_layout Itemize
Common mechanisms of rejection 
\end_layout

\begin_layout Itemize
Effective immune suppression requires monitoring for rejection and tuning
 
\end_layout

\begin_layout Itemize
Current tests for rejection (tissue biopsy) are invasive and biased 
\end_layout

\begin_layout Itemize
A blood test based on microarrays would be less biased and invasive
\end_layout

\end_deeper
\begin_layout Itemize
Memory cells are resistant to immune suppression
\end_layout

\begin_deeper
\begin_layout Itemize
Mechanisms of resistance in memory cells are poorly understood 
\end_layout

\begin_layout Itemize
A better understanding of immune memory formation is needed
\end_layout

\end_deeper
\begin_layout Itemize
Mesenchymal stem cell infusion is a promising new treatment to prevent/delay
 rejection
\end_layout

\begin_deeper
\begin_layout Itemize
Demonstrated in mice, but not yet in primates
\end_layout

\begin_layout Itemize
Mechanism currently unknown, but MSC are known to be immune modulatory
\end_layout

\end_deeper
\begin_layout Subsection*
Overview of bioinformatic analysis methods
\end_layout

\begin_layout Standard
An overview of all the methods used, including what problem they solve,
 what assumptions they make, and a basic description of how they work.
\end_layout

\begin_layout Itemize
ChIP-seq Peak calling
\end_layout

\begin_deeper
\begin_layout Itemize
Cross-correlation analysis to determine fragment size
\end_layout

\begin_layout Itemize
Broad vs narrow peaks
\end_layout

\begin_layout Itemize
SICER for broad peaks
\end_layout

\begin_layout Itemize
IDR for biologically reproducible peaks
\end_layout

\begin_layout Itemize
csaw peak filtering guidelines for unbiased downstream analysis
\end_layout

\end_deeper
\begin_layout Itemize
Normalization is non-trivial and application-dependant
\end_layout

\begin_deeper
\begin_layout Itemize
Expression arrays: RMA & fRMA; why fRMA is needed
\end_layout

\begin_layout Itemize
Methylation arrays: M-value transformation approximates normal data but
 induces heteroskedasticity
\end_layout

\begin_layout Itemize
RNA-seq: normalize based on assumption that the average gene is not changing
\end_layout

\begin_layout Itemize
ChIP-seq: complex with many considerations, dependent on experimental methods,
 biological system, and analysis goals
\end_layout

\end_deeper
\begin_layout Itemize
Limma: The standard linear modeling framework for genomics
\end_layout

\begin_deeper
\begin_layout Itemize
empirical Bayes variance modeling: limma's core feature
\end_layout

\begin_layout Itemize
edgeR & DESeq2: Extend with negative bonomial GLM for RNA-seq and other
 count data
\end_layout

\begin_layout Itemize
voom: Extend with precision weights to model mean-variance trend
\end_layout

\begin_layout Itemize
arrayWeights and duplicateCorrelation to handle complex variance structures
\end_layout

\end_deeper
\begin_layout Itemize
sva and ComBat for batch correction
\end_layout

\begin_layout Itemize
Factor analysis: PCA, MDS, MOFA
\end_layout

\begin_deeper
\begin_layout Itemize
Batch-corrected PCA is informative, but careful application is required
 to avoid bias
\end_layout

\end_deeper
\begin_layout Itemize
Gene set analysis: camera and SPIA
\end_layout

\begin_layout Section*
Innovation
\end_layout

\begin_layout Itemize
MSC infusion to improve transplant outcomes (prevent/delay rejection)
\end_layout

\begin_deeper
\begin_layout Itemize
Characterize MSC response to interferon gamma
\end_layout

\begin_layout Itemize
IFN-g is thought to stimulate their function
\end_layout

\begin_layout Itemize
Test IFN-g treated MSC infusion as a therapy to delay graft rejection in
 cynomolgus monkeys
\end_layout

\begin_layout Itemize
Monitor animals post-transplant using blood RNA-seq at serial time points
\end_layout

\end_deeper
\begin_layout Itemize
Investigate dynamics of histone marks in CD4 T-cell activation and memory
\end_layout

\begin_deeper
\begin_layout Itemize
Previous studies have looked at single snapshots of histone marks
\end_layout

\begin_layout Itemize
Instead, look at changes in histone marks across activation and memory
\end_layout

\end_deeper
\begin_layout Itemize
High-throughput sequencing and microarray technologies
\end_layout

\begin_deeper
\begin_layout Itemize
Powerful methods for assaying gene expression and epigenetics across entire
 genomes
\end_layout

\begin_layout Itemize
Proper analysis requires finding and exploiting systematic genome-wide trends
\end_layout

\end_deeper
\begin_layout Chapter*
1.
 Reproducible genome-wide epigenetic analysis of H3K4 and H3K27 methylation
 in naive and memory CD4 T-cell activation
\end_layout

\begin_layout Section*
Approach
\end_layout

\begin_layout Itemize
CD4 T-cells are central to all adaptive immune responses and memory
\end_layout

\begin_layout Itemize
H3K4 and H3K27 methylation are major epigenetic regulators of gene expression
\end_layout

\begin_layout Itemize
Canonically, H3K4 is activating and H3K27 is inhibitory, but the reality
 is complex
\end_layout

\begin_layout Itemize
Looking at these marks during CD4 activation and memory should reveal new
 mechanistic details
\end_layout

\begin_layout Itemize
Test 
\begin_inset Quotes eld
\end_inset

poised promoter
\begin_inset Quotes erd
\end_inset

 hypothesis in which H3K4 and H3K27 are both methylated
\end_layout

\begin_layout Itemize
Expand scope of analysis beyond simple promoter counts
\end_layout

\begin_deeper
\begin_layout Itemize
Analyze peaks genome-wide, including in intergenic regions
\end_layout

\begin_layout Itemize
Analysis of coverage distribution shape within promoters, e.g.
 upstream vs downstream coverage
\end_layout

\end_deeper
\begin_layout Section*
Methods
\end_layout

\begin_layout Itemize
Re-analyze previously published CD4 ChIP-seq & RNA-seq data 
\begin_inset CommandInset citation
LatexCommand cite
key "LaMere2016,Lamere2017"
literal "true"

\end_inset


\end_layout

\begin_deeper
\begin_layout Itemize
Completely reimplement analysis from scratch as a reproducible workflow
\end_layout

\begin_layout Itemize
Use newly published methods & algorithms not available during the original
 analysis: SICER, csaw, MOFA, ComBat, sva, GREAT, and more
\end_layout

\end_deeper
\begin_layout Itemize
SICER, IDR, csaw, & GREAT to call ChIP-seq peaks genome-wide, perform differenti
al abundance analysis, and relate those peaks to gene expression
\end_layout

\begin_layout Itemize
Promoter counts in sliding windows around each gene's highest-expressed
 TSS to investigate coverage distribution within promoters
\end_layout

\begin_layout Section*
Results
\end_layout

\begin_layout Itemize
Different histone marks have different effective promoter radii
\end_layout

\begin_layout Itemize
H3K4 and RNA-seq data show clear evidence of naive convergence with memory
 between days 1 and 5
\end_layout

\begin_layout Itemize
Promoter coverage distribution affects gene expression independent of total
 promoter count
\end_layout

\begin_layout Itemize
Remaining analyses to complete:
\end_layout

\begin_deeper
\begin_layout Itemize
Look for naive-to-memory convergence in H3K27 data
\end_layout

\begin_layout Itemize
Look at enriched pathways for day 0 to day 1 (activation) compared to day
 1 to day 5 (putative naive-to-memory differentiation)
\end_layout

\begin_layout Itemize
Find genes with different expression patterns in naive vs.
 memory and try to explain the difference with the Day 0 histone mark data
\end_layout

\begin_deeper
\begin_layout Itemize
Determine whether co-occurrence of H3K4me3 and H3K27me3 (proposed 
\begin_inset Quotes eld
\end_inset

poised
\begin_inset Quotes erd
\end_inset

 state) has effects on post-activation expression dynamics
\end_layout

\begin_layout Itemize
Promoter coverage distribution dynamics throughout activation for interesting
 subsets of genes
\end_layout

\end_deeper
\begin_layout Itemize
(Backup) Compare and contrast behavior of promoter peaks vs intergenic (putative
 enhancer) peaks (GREAT analysis)
\end_layout

\begin_deeper
\begin_layout Itemize
Put results in context of important T-cell pathways & gene expression data
\end_layout

\end_deeper
\end_deeper
\begin_layout Section*
Discussion
\end_layout

\begin_layout Itemize
"Promoter radius" is not constant and must be defined empirically for a
 given data set
\end_layout

\begin_layout Itemize
Evaluate evidence for poised promoters and enhancer effects on gene expression
 dynamics of naive-to-memory differentiation
\end_layout

\begin_layout Itemize
Compare to published work on other epigenetic marks (e.g.
 chromatin accessibility)
\end_layout

\begin_layout Chapter*
2.
 Improving array-based analyses of transplant rejection by optimizing data
 preprocessing
\end_layout

\begin_layout Section*
Approach
\end_layout

\begin_layout Itemize
Machine-learning applications demand a "single-channel" normalization method
\end_layout

\begin_layout Itemize
frozen RMA is a good solution, but not trivial to apply
\end_layout

\begin_layout Itemize
Methylation array data preprocessing induces heteroskedasticity
\end_layout

\begin_layout Itemize
Need to account for this mean-variance dependency in analysis
\end_layout

\begin_layout Section*
Methods
\end_layout

\begin_layout Itemize
Expression array normalization for detecting acute rejection
\end_layout

\begin_layout Itemize
Use frozen RMA, a single-channel variant of RMA
\end_layout

\begin_layout Itemize
Generate custom fRMA normalization vectors for each tissue (biopsy, blood)
\end_layout

\begin_layout Itemize
Methylation arrays for differential methylation in rejection vs.
 healthy transplant
\end_layout

\begin_layout Itemize
Adapt voom method originally designed for RNA-seq to model mean-variance
 dependence
\end_layout

\begin_layout Itemize
Use sample precision weighting and sva to adjust for other confounding factors
\end_layout

\begin_layout Section*
Results
\end_layout

\begin_layout Itemize
custom fRMA normalization improved cross-validated classifier performance
 
\begin_inset CommandInset citation
LatexCommand cite
key "Kurian2014"
literal "true"

\end_inset


\end_layout

\begin_layout Itemize
voom, precision weights, and sva improved model fit
\end_layout

\begin_deeper
\begin_layout Itemize
Also increased sensitivity for detecting differential methylation
\end_layout

\end_deeper
\begin_layout Section*
Discussion
\end_layout

\begin_layout Itemize
fRMA enables classifying new samples without re-normalizing the entire data
 set
\end_layout

\begin_deeper
\begin_layout Itemize
Critical for translating a classifier into clinical practice
\end_layout

\end_deeper
\begin_layout Itemize
Methods like voom designed for RNA-seq can also help with array analysis
\end_layout

\begin_layout Itemize
Extracting and modeling confounders common to many features improves model
 correspondence to known biology
\end_layout

\begin_layout Chapter*
3.
 Globin-blocking for more effective blood RNA-seq analysis in primate animal
 model
\end_layout

\begin_layout Standard
\begin_inset Note Note
status open

\begin_layout Plain Layout
Paper title: Optimizing yield of deep RNA sequencing for gene expression
 profiling by globin reduction of peripheral blood samples from cynomolgus
 monkeys (Macaca fascicularis).
\end_layout

\end_inset


\end_layout

\begin_layout Standard
\begin_inset Note Note
status open

\begin_layout Plain Layout
How to integrate/credit sections written by others (e.g.
 wetlab methods)? (Majority of paper text is written by me.)
\end_layout

\end_inset


\end_layout

\begin_layout Standard
\begin_inset Note Note
status open

\begin_layout Plain Layout
Move paper's Background section into thesis Introduction section?
\end_layout

\end_inset


\end_layout

\begin_layout Section*
Approach
\end_layout

\begin_layout Itemize
Cynomolgus monkeys as a model organism
\end_layout

\begin_deeper
\begin_layout Itemize
Highly related to humans
\end_layout

\begin_layout Itemize
Small size and short life cycle - good research animal
\end_layout

\begin_layout Itemize
Genomics resources still in development
\end_layout

\end_deeper
\begin_layout Itemize
Inadequacy of existing blood RNA-seq protocols
\end_layout

\begin_deeper
\begin_layout Itemize
Existing protocols use a separate globin pulldown step, slowing down processing
\end_layout

\end_deeper
\begin_layout Section*
Methods
\end_layout

\begin_layout Itemize
New blood RNA-seq protocol to block reverse transcription of globin genes
\end_layout

\begin_layout Itemize
Blood RNA-seq time course after transplants with/without MSC infusion
\end_layout

\begin_layout Section*
Results
\end_layout

\begin_layout Itemize
New blood RNA-seq protocol increases effective yield 2-fold while maintaining
 sample quality (paper)
\end_layout

\begin_layout Itemize
MSC treatment signature is swamped by much larger post-transplant stress/injury
 response (analysis to demonstrate application of developed protocol to
 real data)
\end_layout

\begin_layout Section*
Discussion
\end_layout

\begin_layout Itemize
Globin-blocking is highly effective and efficient for blood RNA-seq
\end_layout

\begin_layout Itemize
More work required to tease out subtle post-transplant MSC signature in
 living animals
\end_layout

\begin_layout Part*
Future Directions
\end_layout

\begin_layout Itemize
Study other epigenetic marks in more contexts
\end_layout

\begin_deeper
\begin_layout Itemize
DNA methylation, histone marks, chromatin accessibility & conformation in
 CD4 T-cells
\end_layout

\begin_layout Itemize
Also look at other types lymphocytes: CD8 T-cells, B-cells, NK cells
\end_layout

\end_deeper
\begin_layout Itemize
Investigate epigenetic regulation of lifespan extension in 
\emph on
C.
 elegans
\end_layout

\begin_deeper
\begin_layout Itemize
ChIP-seq of important transcriptional regulators to see how transcriptional
 drift is prevented
\end_layout

\end_deeper
\begin_layout Standard
\begin_inset ERT
status open

\begin_layout Plain Layout

% Use "References" instead of "Bibliography" 
\end_layout

\begin_layout Plain Layout


\backslash
renewcommand{
\backslash
bibname}{References}
\end_layout

\end_inset


\end_layout

\begin_layout Standard
\begin_inset CommandInset bibtex
LatexCommand bibtex
bibfiles "refs"
options "plain"

\end_inset


\end_layout

\end_body
\end_document