Flags and Lollipops Open Lab Notebook
- http://www.ghastlyfop.com/blog/2007/12/open-notebook-pt1.html
- http://www.ghastlyfop.com/blog/2007/12/open-notebook-pt2-question-theories.html
- http://www.ghastlyfop.com/blog/2007/12/open-notebook-whats-disease-again.html
Alison Meynarts MSc Research
- Used GeneRIFs, gene2pubmed, Swissprot PubMed Links
Combining P-values (Fisher's Method derivation by Lou Jost)
- http://www.loujost.com/Statistics and Physics/Significance Levels/CombiningPValues.htm Copy and paste the link
Also investigate TOPPGENE
- http://www.biomedcentral.com/1471-2105/8/392/abstract
- Use Mammalian Phenotype Ontology (this has links to mouse genes?)
- Assume orthologous genes implies similar phenotype
- use co-occurrence of genes in PMID via gene2pubmed as implying gene interaction
- Compared to SUSPECTS and ENDEAVOR
- Used "information content" - occurrences of term and children vs all terms
- Uses Sugeno "Fuzzy Measure"
- Should see if I can extract validation set from supplementary files
Also extract algorithms covered in Proposal
Links to Related Gene Prioritization Algorithms
(courtesy of Gary Bader)
Association of genes to genetically inherited diseases using data mining.
http://www.nature.com/ng/journal/v31/n3/abs/ng895.html
G2D (See proposal)
The human disease network.
http://www.pnas.org/cgi/content/full/104/21/8685
Analysis of human disease network obtained from OMIM.
A human phenome-interactome network of protein complexes implicated in
genetic disorders.
http://www.nature.com/nbt/journal/v25/n3/abs/nbt1295.html
Interactome-Phenome network. Assume mutations in interaction partners causing phenotype can be projected to interaction partners.
Gene prioritization through genomic data fusion.
http://www.nature.com/nbt/journal/v24/n5/abs/nbt1203.html
*this one comes with software called Endeavour, which claims to be easy
to use*
Endeavor, information fusion for gene prioritisation.
A computational system to select candidate genes for complex human traits
http://bioinformatics.oxfordjournals.org/cgi/content/full/23/9/1132
CAESAR system
Genome-wide identification of genes likely to be involved in human
genetic disease
http://nar.oxfordjournals.org/cgi/content/full/32/10/3108
Prediction from sequence properties
Discovering disease-genes by topological features in human
protein–protein interaction network
http://bioinformatics.oxfordjournals.org/cgi/content/abstract/22/22/2800
PPI to build KNN classifier for prediction (print and read!)
Analysis of protein sequence and interaction data for candidate disease
gene prediction
http://nar.oxfordjournals.org/cgi/content/abstract/34/19/e130
Computational disease gene identification: a concert of methods
prioritizes type 2 diabetes and obesity candidate genes
http://nar.oxfordjournals.org/cgi/content/abstract/34/10/3067
Multiple methods applied to the same problem
SUSPECTS: enabling fast and effective prioritization of positional
candidates
http://bioinformatics.oxfordjournals.org/cgi/content/abstract/22/6/773
Target SNP selection in complex disease association studies
http://www.biomedcentral.com/1471-2105/5/92
POCUS: mining genomic sequence annotation to predict disease genes
http://genomebiology.com/2003/4/11/R75
Comparative Toxicogenomics Database
http://ctd.mdibl.org/
- Database of curated/inferred chemical/gene/disease/pathway interactions (pairwise)
- Entrez/Unigene, OMIM, KEGG, MeSH
- Inference via transitivity (chem-gene + gene-disease = chem-disease)
TCMGeneDIT: a database for associated traditional Chinese medicine, gene and disease information using text mining
http://www.biomedcentral.com/1472-6882/8/58/abstract
- Trad. Chinese Medicine/gene/disease, using MeSH terms for disease.
- Transitivity (Swanson ABC) - our profile score is like doing the Swanson ABC en masse.
- PharmGKB (gene/disease/outcome database) - stores disease outcomes preferably as MeSH terms
GeneCards Disease Genes List
http://www-bimas.cit.nih.gov/cgi-bin/cards/listdiseasecards.pl?type=full
Gene to Expression Data (with MeSH term annotation)
http://www.cleanex.isb-sib.ch/
http://cbrc.musc.edu/homepage/jani/genemesh/about.html
GeneMeSH does hypergeometric probability linking of Genes-to-MeSH via gene2pubmed
GenoMeSH (http://www.genomesh.org/ He Group)