1.1 Binary file budgetJust.pdf has changed

2.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000 2.2 +++ b/budgetJust.txt Fri Apr 24 01:12:36 2009 -0700 2.3 @@ -0,0 +1,54 @@ 2.4 +\documentclass[11pt]{nih-blank} 2.5 + 2.6 + 2.7 +\begin{document} 2.8 +== Budget Justification == 2.9 + 2.10 +Because the anticipated product is mainly intellectual property (approaches for extracting information from 2.11 +large data sets of a particular sort), this work depends most heavily on personnel, as reflected in the budget. 2.12 + 2.13 +\vspace{0.3cm} \noindent **A. Senior/Key Person** 2.14 + 2.15 +\noindent 1. Charles F. Stevens, M.D., Ph.D., Principal Investigator (effort \= 4.80 calendar months), will direct the 2.16 +project, engage in improving the computational approaches from the ones currently being used, contribute 2.17 +to developing the mathematical basis for the algorithms, participate in the interpretation of data, and write 2.18 +drafts of journal articles. Stevens is known for combining theory with experiment, is trained in physics and 2.19 +mathematics as well as in biology, and has been using computation approaches to data analysis for many 2.20 +years in many different contexts. 2.21 +\\\\ 2.22 +In the subsequent year a 3% increment has been calculated. 2.23 + 2.24 +\vspace{0.3cm} \noindent **B. Other Personnel** 2.25 + 2.26 + \indent Postdoctoral Associates. 2.27 + 2.28 + To be named, Ph.D., Research Associate (effort \= 12 calendar months), trained in computational 2.29 +techniques and machine learning to work collaboratively on the problem of deriving cortical arealization from 2.30 +gene expression patterns. 2.31 + 2.32 + Graduate Students. 2.33 + 2.34 + Baylis Shanks, M.S. (effort \= 12 calendar months), is currently working on the project for his thesis. 2.35 +Shanks is currently in the Computational Neuroscience program at the University of California, San Diego, 2.36 +and has been primarily responsible for applying the suite of standard approaches to the problem of 2.37 +identifying genetic tags that identify cortical areas. He is well trained in computational methods for the 2.38 +analysis of large data sets, and is at home with neurobiology. 2.39 + 2.40 +\vspace{0.3cm} \noindent **D. Travel** 2.41 + 2.42 +\noindent 1. Domestic. Funds are requested to cover the cost of attending three scientific meetings to present work 2.43 +related to this project at \$1,500 each (airfare, ground transportation, registration, lodging, meals). 2.44 + 2.45 +\vspace{0.3cm} \noindent **F. Other Direct Costs** 2.46 + 2.47 +\noindent 1. Material and Supplies. Computer related consumables, software licenses \$11,000; computer 2.48 +equipment, including mass storage \$6,000. 2.49 +\\2. Publication costs. Graphics, page charges, and reprints. 2.50 +\\4. ADP/Computer Services. Institutional computer services and network access charges. 2.51 +\\8. Central Services. Institutional services for which investigators are charged on a fee-for-use basis 2.52 +include lab support shop, small equipment repair and maintenance, photocopying, and postage. 2.53 +\\9. Graduate Student Fees. The University of California, San Diego resident fees for 2009-2010 are 2.54 +currently estimated at \$10,945, and a portion (\$5,468) is budgeted so that in year 1 the combined 2.55 +salary/benefits/fees amount does not exceed the maximum amount allowed by NIH (\$37,368). 2.56 + 2.57 +\end{document}

3.1 Binary file grant.pdf has changed

4.1 --- a/grant.txt Thu Apr 23 03:12:36 2009 -0700 4.2 +++ b/grant.txt Fri Apr 24 01:12:36 2009 -0700 4.3 @@ -205,7 +205,7 @@ 4.4 \cite{thompson_genomic_2008} describes an analysis of the anatomy of 4.5 the hippocampus using the ABA dataset. In addition to manual analysis, 4.6 two clustering methods were employed, a modified Non-negative Matrix 4.7 -Factorization (NNMF), and a hierarchical recursive bifurcation clustering scheme based on correlation as the similarity score. The paper yielded impressive results, proving the usefulness of computational genomic anatomy. We have run NNMF on the cortical dataset 4.8 +Factorization (NNMF), and a hierarchical bifurcation clustering scheme using correlation as similarity. The paper yielded impressive results, proving the usefulness of computational genomic anatomy. We have run NNMF on the cortical dataset, and while the results are promising, other methods may perform as well or better (see Preliminary Studies, Figure \ref{dimReduc}). 4.9 4.10 %% \footnote{We ran "vanilla" NNMF, whereas the paper under discussion used a modified method. Their main modification consisted of adding a soft spatial contiguity constraint. However, on our dataset, NNMF naturally produced spatially contiguous clusters, so no additional constraint was needed. The paper under discussion also mentions that they tried a hierarchical variant of NNMF, which we have not yet tried.} and while the results are promising, they also demonstrate that NNMF is not necessarily the best dimensionality reduction method for this application (see Preliminary Studies, Figure \ref{dimReduc}). 4.11