Here is an example of a specific aims page from a funded grant of mine from 2004. This was funded as-is on the first round of submission. It received a score of 145, which at the time was 6th percentile.
Note that this is not a very long specific aims page. It is not long, because it didn’t need to be. The key point is that it was focused and clear. By reading this, reviewers were able to tell what I wanted to do, why I wanted to do it, and how I was going to accomplish it. They could also see why we were the right people for the job. These are all key elements of a specific aims.
D. Research Plan
Specific Aims
Present proteomics software is limited to identifying and characterizing mainly proteins for which a gene or protein entry exists in one of the public databases. Protein identification cannot be effectively performed for organisms whose annotations are incomplete, missing, or incorrect. We have developed new software called Genome Fingerprint Scanning (GFS) that is capable of matching mass spectrometry (MS) data from proteomic studies directly to raw (even unfinished) genome sequence, identifying the coding locus for an observed protein[1]. The program has been used to identify novel proteins in Tetrahymena thermophila and Francisella tularensis, as described in the preliminary results. We have also developed a preliminary website for end users, at gfs.unc.edu.
The aim of this proposal is to transform GFS from an experimental, beta-quality tool, into a free, widely-used community resource that can be accessed either through a web interface or through local installation on the researcher’s own computers. We propose the following specific aims:
- Develop a professional-quality website for end users of GFS. This will include greatly enhancing the program output to include peptide maps that users can browse overlaid on a genome; allowing users to submit any sequence against which to search MS data; greatly expanding the list of built-in searchable genomes on the website; providing a multi-genome simultaneous search capability; automatically updating the genome databases from their respective sources; automating distribution of the computing load for website requests on our cluster to ensure rapid response time; and sequentially revising the site to address ease-of-use, based on user surveys.
- Enhance the performance of GFS and extend applicability to large genomes. We plan to enhance the matching process for multi-exon genes, providing use with human and other complex genomes to identify proteins; to develop a probability-based scoring function providing immediate feedback regarding the probability of a false-positive match; to significantly enhance the tandem mass spectrometry (MS/MS) peptide matching feature; to greatly optimize the speed of GFS for use with large genomes, using b-tree indexing and disk caching; to provide automatic program parameter adjustment to obtain the optimal results; and to perform thorough testing on the code as well as real-world testing of the program with standards data, making the standards data also available to the community.
- Port the code and develop documentation. We will port the code to Windows and common Unix platforms such as Linux, providing both executable and source code versions; and develop thorough documentation aimed at all user levels, including developers, administrators, and end-users.