8 Structure of a Scientific Research Proposal
In Biocore lab you will write research proposal papers before you collect data for your research projects. A research proposal is a very important first step that helps you get familiar with your system and serves as a guide for your entire project. The proposal has many similar attributes as a lab paper (discussed in the previous section) and shares nearly all the same components; Introduction, Materials and Methods, and Expected & Alternative Results. We call the final section “Implications” rather than a “Discussion” to emphasize the potential impact of the predicted expected and alternative results.
Below we point out how proposals differ from final lab papers and provide guidelines for what should be included in this type of paper. When writing about what you propose to do, use the future tense. No abstract is necessary for research proposals.
Make sure to review the Research Proposal Rubric as you are writing!!
Title
See Final Lab Report Title section description and examples. How does a proposal title compare to a title for a final paper? Compare rubrics.
Introduction
Include a summary of background information, experimental question, biological rationale, hypothesis, and experimental approach. As you become more familiar with your system during your study, you will likely need to revise this section for your final paper to reflect the greater depth of your knowledge or unanticipated variables that become clear as the study progresses.
Methods
The methods section is usually quite detailed and may include diagrams or flow charts explaining your experimental design and protocols. Include a description of any pilot studies you plan to do.
Expected and Alternative Results
Since you have not done the experiment yet, you will not have any data. However, your hypothesis is a clear statement of what you expect and should provide the basis for this section. Provide a graph of the data you expect if your hypothesis is supported, showing actual numbers on labeled axes. This data is ‘dummy data’ – you make it up to represent expected trends and variation based on your current knowledge of the system. It could be based on your own pilot studies and/or published data from similar studies. Text accompanying this section should point out expected trends and describe pertinent attributes of trend lines. You should also present biologically plausible alternative results to those you expect, e.g., opposite results or the “no difference” result. (Do not present alternative results that represent flawed mechanical assumptions.) Thinking about alternative results at the proposal stage may help you troubleshoot problems, evaluate the efficacy of your control, or provide a background for your final results since, quite often, these are the ones you actually see at the end of your experiment.
Implications & Conclusions
In this section, describe the implications of the predicted trend described in your expected results as it relates to the knowledge gap and the broader rationale presented in your Introduction. Remind the reader of the biological and methodological assumptions you are making, and limitations of your experiment. Discuss your alternative results and explain how they might yield from incomplete or alternative rationale or unanticipated variables. Describe limiting factors (e.g. replication, controls etc.), and evaluate your confidence in the experimental design and/or your capacity to make broad conclusions. Finish off with a strong conclusion, with a description of ramifications if your hypothesis is supported. Special note on avoiding social justifications: You should not over emphasize the relevance of your experiment and the possible connections to large-scale processes. Be realistic and logical—do not over generalize or state grand implications that are not sensible given the structure of your experimental system. Not all science is easily applied to improving the human condition (cure cancer or solve climate change). Performing an investigation just for the sake of adding to our scientific knowledge (“basic science”) is important too. In fact, basic science often provides the foundation for applied studies.
Example of Good Implications
Adapted from a paper by Claire Evensen- Biocore 382 Fall 2017
Inoculation of Solidago canadensis with rust fungus expected to result in higher infection severity on younger, upper leaves as compared to older, lower leaves
Knowledge Gap: Although it is known that rust fungus infects S. canadensis leaf tissue (Novander and Smith 1995), it is not known if infection severity is influenced by leaf age, nor is it know if the age gradient across a single plant from older leaves on the lower stem to younger leaves on the upper stem is significant enough to result in differences in infection severity.
Implications:
If we see differences in infection severity between older and younger leaves treated with a fungal spray, the study will support the idea that stomatal opening arising from differences in leaf age is an important factor in rust fungus infection. Though it was previously know that infections occur via the stomata, it was unclear whether the variation in stomatal opening associated with leaf age was distinct enough to either hinder or advance the fungal infection process. Although we are not measuring the degree of stomatal opening or closure, if we support our hypothesis that younger leaves are more susceptible to infection than older leaves, our data would suggest that the age of leaf including lack of open stomata more prevalent in older leaves dramatically lowers the probability of the fungal germ tube finding an insertion site—to the point that a large proportion of spores that adhere to the leaves are unsuccessful in entering the host tissue (Bradley et al. 2007). An alternative explanation for higher infection rate on younger leaves is simply associated with stem height, with fungal spores more easily spread by wind to leaves that are higher on the plant stem as opposed to older leaves that are less exposed and lower on the stem (Novander and Smith 1995). Regardless of the mechanism, our work will provide valuable insight into how the relationship between the rust fungus and S. canadensis changes with leaf age. (Referring back to biorationale and comparing expected results with knowledge gap).
If our study yields alternative results and we reject our hypothesis, we could conclude that either our assumptions regarding stomata opening and age are flawed or there are unanticipated confounding factors influencing our study. We assume that older leaves would have fewer stomata openings and would, therefore, provide fewer opportunities for fungal infections in the older leaf tissue. However, if the rust fungus germ tube is highly efficient in terms of leaf coverage, or if a robust infection only requires a baseline threshold of a “few” stomata, and if there are enough stomata available to be sufficient for infection even when a leaf has almost completely senesced, we will likely not see statistically significant differences in infection between younger and older leaves. An important additional variable includes the presence of prior infections. In other words, we may spray plants that were already infected with spores that had yet to germinate. Should this occur, statistically significant differences in infection rate? may be masked by a previous rust infection. (Explaining how assumptions, unanticipated variables, and limiting factors, here and below, could yield alternative results)
Our study is limited by our inability to control the presence of naturally occurring wind-borne rust spores. We assume that a single wind-borne spore has a low probability of adhering to a S. canadensis leaf, however it is possible for natural infection to contaminate and obscure potential differences due to our inoculation treatments. This experiment will be done in a field setting in the Biocore Prairie. As such, we have located a patch of S. canadensis with no apparent infection that is isolated from infected patches of other S. canadensis plants. We will be creating a spray inoculant at saturating concentration and at a much higher concentration than naturally occurring spores could achieve. Both the treated and control plants will be isolated by dense vegetation and therefore, will be much more likely targets of infection by our treatments than by natural infection. If there is any contamination by naturally occurring wind-borne spores, we will detect it on our control group’s extent of infection following the experiment. The extent of infection by non-inoculated control plants will serve as a baseline for comparison to the two treatment groups. (Reminding the reader of the biological and methodological assumptions you are making, and limitations of your experiment.)
Finally, we assume that the Tween-20 solution will be a suitable temporary environment for spores. Rust fungus is highly dependent on its relationship with its host plant (Petersen 1974), so it may be weakened or die when it is removed from the host. Should this occur, we will expect to see low levels of infection across all three groups, as manual infection attempts would fail. Nevertheless, we are confident in our design given the timing of our study in mid-Sept when the life cycles of both the host and the fungus align; the ideal germination temperature for the fungus of 37°C will be achieved; and that previous studies have found success with the 0.01% Tween-20 solution (Stavely 1983). (Evaluation of confidence in method)
In conclusion, we believe our rationale regarding stomatal infection mechanism, and the relationship of stomatal opening and leaf age is sound. Although there is literature describing the mechanism of rust fungal infection through stomatal opening, to our understanding, it is not established that infection by the S. canadensis leaf rust fungus is associated with leaf age. If our hypothesis regarding leaf age of S. canadensis and rust infection severity is supported, we can better predict incidence and timing of rust infection on S. canadensis and can furthermore, support questions about control and spread of S. canadensis and this fungal leaf pathogen. (Ramifications if hypothesis is supported)
