| |
Legacy data and other
collections
In
the context of the present proposal, the MBL is exclusively produced
in-house. However, our approach is generic, and in the future we plan to
incorporate material from other sources. For example, the entire
Yakovlev-Hakem collection of human brains at the Armed Forces Institute of
Pathology, which were also processed in celloidin and cut at 35 m, is an
obvious candidate. Relaxation of the strict protocols for tissue
processing and data collection imposed in Project 1 will
initially
limit the scope of material that can be segmented with accuracy. Yet we
believe that modification of the algorithms will eventually make it
possible to incorporate a wider range of data sets without compromising
the precision of segmentation. Certainly at this stage, accommodation of
legacy data can made at the level of anatomical homology. Indeed, this
mode of linkage may have the greatest repercussions in neuroscience and
clinical research. At this level we may provide a gateway to genomics
databases for other vertebrate brain phenotyping efforts. For example,
consider the following scenario: An abnormality is discovered in a
functional neuroimaging study (e.g., Gur et al. 1995) and indicates a
serious structural abnormality (e.g., Arnold et al. 1995). If variations
in the corresponding structures have already been identified in the MBL,
then it may be possible to establish a link between relevant QTLs and
genes in mouse that may underlie the abnormality in humans. Anatomical and
genomic homologies are essentially cross-referenced.
|
|
