Donald Doherty's blog

Tag: Brain Science

  • The Database of Odor Responses (DoOR): a Functional Atlas of All Available Odor Responses

    Odors are recognized by a large family of odor receptors. Each receptor cell expresses one or a few receptor proteins, which give that cell a specific odor response profile. This profile can be represented by a mathematical function.

    When we smell, most chemicals stimulate more than 1 odor receptor cell type. The result is that each odor elicits an activity pattern across an array of odor receptor cells. Population coding enables the brain to recognize and remember thousands or maybe millions of different odors with a limited number of receptor types (approximately 350 in humans and about 60 in fruit flies).

    The recent paper “Integrating Heterogeneous Odor Response Data into a Common Response Model: A DoOR to the Complete Olfactome” (published September 2010 in Chemical Senses) reports on the creation of a functional atlas of odor responses for not only odor receptors but also the olfactory glomeruli of the fruit fly.

    The project is important for two reasons:

    • The functional atlas represents a consensus data set combining all available data that can serve as a standard reference for the sense of smell in fruit flies. Since the odor response profiles are based on many studies they are statistically more reliable than any single study.
    • The computer software tools test an approach for mapping different data sets onto each other.

    The team developed a software platform that allows the extraction of odor response profiles across chemicals for individual receptors or the extraction of the entire combinatorial response pattern elicited by a given chemical. The software is an open source R-package and can be downloaded from the DoOR website. The download also includes the fruit fly data.

    Note: R-packages are written in the R programming language and and must be run in the R software environment for statistical computing and graphics. The software is all open source and freely available through the R Project for Statistical Computing website.

  • NIF: Better Literature Search

    Over the past couple of days we’ve looked at the data aggregating capabilities of the Neuroscience Information Framework (NIF). What about literature search? Is there any reason you should move from PubMed or Google Scholar to NIF?

    Figure 1. NIF literature search results for “barrel cortex.” NIF provides four tabs on the literature search results page.

    NIF provides the results of a standard PubMed search under the PubMed tab. In addition, NIF provides Open Access Literature, Neuronal Morphology, and Neuroscience Literature tabs. The Open Access Literature tab provides a convenient way to go directly to relevant articles that are freely available over the Internet. The Neuronal Morphology tab provides quick access to papers associated with digitally reconstructed neurons available through NeuroMorpho.org. The Neuroscience Literature tab lists papers returned based on a full text search using your search words. This could arguably the most valuable additional literature search service that NIF provides.

    Unfortunately the majority of papers are still published in journals that are not open access. Because of legal restrictions, NIF is only able to provide full text search services on a subset of papers that include those published in open access journals and those published in the Journal of Neuroscience. NIF provides a complete list of journals searched here.

    There is a large literature surrounding the amyloid beta protein. Recent evidence suggests that amyloid beta 42 is particularly important in Alzheimer’s disease. What if we only want papers that used amyloid beta 42 in their research? I decided to run a quick test of the full text search capabilities by typing the following “amyloid beta 42” into the NIF home page search box and clicking on the search icon. That didn’t even kick off a search but took me to a blank NEUROLEX page (under neurons and brain regions). The same thing happened when I typed in “amyloid-beta 42” but “abeta42” worked.

    Note: All of the variants above worked when I typed them into the search box returned by the initial search from home page (like at top left in Figure 1 above). Surprisingly, searches for “amyloid beta 42” and “amyloid-beta 42” each resulted in zero hits under “Neuroscience Literature.”

    A search for “abeta42” returned 161 papers under Neuroscience Literature. I checked all the articles in the first three pages and the last page and they all included abeta42 in the title or the abstract so I’m not sure if we’ve located additional articles that, for example, may have only mentioned the 42 amino acid peptide in its methods section. Also, it doesn’t look like synonyms are being used in the search.

    Other related blog posts:

    NIF: When You’re Looking for Neuroscience Resources Including Data

    NIF: Neurons, Models, and Grants

  • NIF: Neurons, Models, and Grants

    Yesterday we took an initial peek at the Neuroscience Information Framework (NIF). Today I’d like to briefly point to the range of Data Federation resources you can find there. When “barrel cortex” is typed into the NIF search box and a search is executed, a couple of subcategories come up under the Nervous System Level category and about five subcategories under Data Type like in Figure 1 below.

    Figure 1. NIF search results for “barrel cortex.” The search words refer to whisker related structures in layer 4 of somatosensory cortex.

    In Figure 1 above, the NeuroMorpho: NeuronInfo entry under the Cellular Level subcategory is selected which results in showing a list of some of the 66 neurons from the NeuroMorpho.org repository that are associated with the barrel cortex.

    Figure 2. A section from a mouse brain in BrainMaps.org. The barrel cortex is labeled “barrels” in red with yellow background.

    Click on the plus (+) sign to the right of Brain Regions to display entries in that subcategory. The BAMS: BrainRegions entry is associated with an anatomy ontology that provides standard naming conventions for brain structures. The BrainMaps:Atlas entry provides access to relevant high-resolution histology sections. Click on the barrel cortex, barrels link under Brain Region and the BrainMaps.org atlas will appear like in Figure 2 above.

    Under the Data Type category, the Images subcategory lists the sites we’ve already looked at. The Models subcategory lists computational model sites relevant to our search. The ModelDB: Models entry lists three models of barrel cortex in the SenseLab ModelDB repository.

    You may be puzzled why the Grants subcategory under Data Type shows large number of hits (73 while writing this) but the Grants tab at the top of the section has none. The Grants subcategory provides access to resources on past and currently funded grants. In contrast, the Grants tab provides resources for those wanting to apply for federally funded grants.

    Yesterday we looked at resources available under the Microarray subcategory. Our current search brought up 1 resource at the Gene Expression Omnibus (GEO). Finally, under Connectivity you will find information on the known neural connections to and from the brain area or structure you searched on. Tomorrow we’ll take a look at what the NIF Literature tab provides.

    Other related blog posts:

    NIF: When You’re Looking for Neuroscience Resources Including Data