proposal for a evidence code

[Judy Blake]

This sounds like mapping data to me...mapping phenotypes in various 
organisms via shared genetic markers/polymorphisms.  The mapping data would 
have it's own representation and evidence in your information 
system...maybe handled in some similar fashion as the GO, maybe not.

The association of the GO term with the gene product would be via a 
separate set of experiments that measure function or some biological 
process.  Inferred from mutant phenotypes (IMP) is one of the evidences 
that could be used there.

I think, then, that the association of the GO term with the gene product 
would be in the primary species.  So, a rice gene product may be shown to 
have an involvement in 'timing of seed set'.  The location of that gene in 
the other cereal genomes would be mapped using the techniques below.  Then 
the researcher might decide to transfer the GO association with the rice 
gene product to the other cereal gene product.  Then the evidence for the 
association of the GO term in the cereal species would perhaps be 'inferred 
from electronic annotation' since you have 1) established the GO 
associations with the primary species by whatever GO evidence code, 2) 
established a relationship between the primary species and the secondary 
species by some other experimental evidence, either by mapping phenotypes 
or sequence similarity or other, and 3) transferred the GO information from 
the primary species to the secondary species based on the identity 
established between the two that is independent of the original GO 
assignment.

I remember working through this type of thing when we discuss the 
transference of information from mouse to rat via orthology determinations 
between the two species.

Judy


At 03:33 PM 11/8/01, Pankaj Jaiswal wrote:

>Dear GO Friends,
>
>
>In Gramene (www.gramene.org), we are presently curating rice 
>mutants/phenotypes
>that have been characterized using
>one or more of the techniques outlined in the list of Evidence codes.  We are
>proposing the
>introduction of a new line of evidence called, "IAGP: Inferred from 
>Association
>between Genotype
>and Phenotype".
>
>We could have used the GO code, IGI (inferred from genetic interaction). 
>However
>after looking
>carefully at the description of IGI, we felt that by using it we might not
>justify the evidence type. The
>reason we are proposing this new code of evidence is because many of the
>phenotypes in crop plants
>(in general from cereal crops ) are characterized using genetic markers and
>populations or genetic
>stocks that are well known to plant geneticists and breeders, and we felt that
>listing them
>specifically will help us clarify the nature of evidence used to define the
>relationships between
>the phenotypes and the genotypes, as defined by markers on the genetic and
>physical maps.
>
>   This is going to be very important while curating information related to
>functional and comparative
>   genomics. Entries in the rice Mutant database will use primarily the GO 
> terms
>and Trait terms (TO) for
>   defining both the morphological and physiological (Biochemical)
>mutants/phenotypes, and the
>   Evidence code would include the following in addition to the presently 
> defined
>ones:
>
>
>   PROPOSED NEW EVIDENCE CODE :
>
>   IAGP  (inferred from association between genotype and phenotype)
>
>   #       Based on detection of polymorphism or segregation of genetic 
> markers
>e.g.. isozymes, RFLPs
>   (Random Fragment Length Polymorphism), RAPDs (Random amplified polymorphic
>DNA), AFLPs
>   (Amplified Fragment Length Polymorphism), SNPs (Single Nucleotide
>Polymorphisms), Microsatellite markers or SSR (Simple Sequence Repeats), TD
>   (Transposon Display).
>
>   #       Based on detection of polymorphism or segregation of physical 
> markers
>e.g.. FISH,
>   centromeric,heterochromatic regions, chromosomal banding patterns.
>
>   #       Based on detection of polymorphisms in segregating plant material
>derived from bi-parental
>   crosses e.g.. F2 lines, F3 families, Back cross populations, viz., BC1, BC2
>etc.; Doubled Haploid lines
>   (DH), Recombinant Inbred Lines (RIL).
>
>   #       Based on detection of polymorphisms in genetic stocks, e.g., Near
>Isogenic Lines (NIL),
>   Introgression Lines (IL), Radiation Hybrids (RH), Cytogenetic Stocks (CG),
>i.e., trisomics, aneuploids,
>   etc.
>
>
>   We look forward to getting your feedback on this proposal.
>
>Sincerely
>
>Pankaj
>
>
>**************************************************************
>Pankaj Jaiswal, Ph.D.
>Postdoctoral Associate
>Dept. of plant Breeding
>Cornell University
>Ithaca, NY-14853, USA
>
>Tel:+1-607-255-3103 / Fax:+1-607-255-6683
>E mail: pj37 at cornell.edu
>http://www.gramene.org   http://ars-genome.cornell.edu/rice
>**************************************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://brie4.cshl.edu/pipermail/gramene/attachments/20011109/d53a2165/attachment.html>