Transcript Under Construction” - University of Toronto

Assembling Your
Specific Aims
February 28, 2013
Specific Aims

Should address:
– What are the major questions your
project is designed to address?
– What are the objectives of your project?
– How will each objective address the
project-specific questions?
Building a Specific Aim
Aim title
“To determine”/”Determining” X in Y
- Use strong action verbs where possible
- Don’t hand wave or write “diffidently”
“To characterize the effect of Drug X on mitochondrial activity in
acute leukemia”
Vs.
“To understand the mechanism of action of Drug X.”
Vs.
“Drug X and metabolism”
“Transcriptional profiling of Drug X”
Building a Specific Aim
Rationale
“Given that [X, Y and Z], it is likely
that [A, B and C] are true”
- Set up as a series of logic statements
- It’s OK to recap some preliminary data or
language from the global rationale
(expected, even)
- Can be combined with your hypothesis
statement, but doesn’t have to be – and in
some cases, probably shouldn’t be
Building a Specific Aim
Hypothesis
“We therefore hypothesize X”
- Make sure it’s falsifiable!
“Drug X impairs mitochondrial activity by interfering with oxidative
phosphorylation”
Vs.
“Drug X affects mitochondria.”
Vs.
“Mitochondrial activity is altered in AML”
Building a Specific Aim
Study
design
“In order to test this hypothesis, we
will use samples derived from [X and
Y] in order to conduct experiments [1,
2, 3 and 4]”
- Don’t go overboard with methods –
unless it’s not something the reviewer is
expected to be familiar with
- Outline where your sample tissues are
coming from
- Give enough detail so that the reviewer
knows what you intend to do
- For discovery grants: Give specifics of
data analysis methodology here
- Make sure your study design is SMART
Study Design vs.
Methodology

Describe your experiments in detail

Focus on study design
– Study design should match up with objective

Reference previously established
techniques, except for specific modifications
– For example, no need to describe RNA isolation
in detail…unless that’s your project!

Need to address where your samples are
coming from
Building a Specific Aim
Anticipated
Outcome
“If [X] is observed…”
- Be CONCISE yet SPECIFIC
- Don’t do more than one outcome at a
time, as a rule – otherwise, it makes it hard
to follow what’s going on.
“We anticipate that oxidative phosphorylation is inhibited in a
dose- and time-dependent manner upon treatment with Drug X.”
Vs.
“If Drug X kills cells by shutting down oxidative phosphorylation….”
Building a Specific Aim
Interpretation
“…then [Y] must be true”
- Directly linked to your anticipated
outcome
- Your chance to explain what you think
your outcomes mean and how they
support/refute your hypotheses
Building a Specific Aim
Contingency “However, if A is observed, then we
will do B”
-Your chance to say “OK, I know things can
go wrong – here’s why this isn’t a lost
cause!”
- Contingencies can be necessary for
technical reasons as well as for scientific
ones
- Repeat
outcome/interpretation/contingency
IF/THEN statements as often as necessary
Building a Specific Aim
Summary
and
significance
“Therefore, taken together, these
studies will allow us to determine [X]”
- Feel free to restate your aim language
- Pull it all together – what does this aim
mean, taken as a whole?
Caveats

Mahadeo’s favourite way to write aims

NOT the only approach
– Personal idiosyncrasies in style
– Clinical aims usually don’t fit the model
– Discovery aims can fit the model, with
some creativity
How to use
Figures/Tables





Show off somebody else’s preliminary
data
Schematics of models
Workflows (especially for data analysis
methods)
Timelines
“Concept maps”
QUESTIONS?