Research Blog
Archive for July, 2010
FANMOD et al.
Jul 20th
FANMOD
Size 3 Subgraph (100,000 Samples, 100 Random Network)
- Approximate number of subgraphs: 265326840 (based on 100000 samples)
- 172022813 subgraphs were enumerated in the original network.
- Result File
Size 6 Subgraph (100,000 Samples, 100 Random Network)
- Approximate number of subgraphs: 9365924517517080 (based on 100000 samples)
MFINDER1.2
3 (Default Settings)
- -
6 (Default Settings)
- -
Overrepresentation
Jul 19th
Stats
Jul 13th
Motif A (Chlorpromazine):
- There are 86 Edges and 76 Nodes in total in the neighbourhood for this motif. (162)
- The similarity between the compound Chlorpromazine and Trimeprazine has a 0.85 Tanimoto co-efficient.
- There are 60 other compounds that bind to the H1 Histamine Receptor.
- Chlorpomazine has 3 other similar drugs. (Tanimoto is 0.85 for all)
- Trimeprazine is similar to 4 additional compounds. Tanimoto betweeen 0.87 – 1.0.
- The H1 Histamine Receptor has 1 additional target and is associated with 2 proteins one of which has a similar structure.
Motif A (Lenalidomide)
- Total of 21 Nodes and 25 Edges in this motif. (46)
- Lenalidomide has 1 additional target.
- Talidomide has 2 additional targets.
- 0.85 is the tanimoto coefficient in both directions if Lenalidomide and Talidomide.
- TNF has 14 other drugs that bind to it and 1 Protein.
Motif C (Isradipine)
- Blast score between CACH2 and SCN4A is 2.0E-63.
- Blast score in the reverse direction is 1.0E-59.
Motif Examples
Jul 12th
Looked at results for some good “known” off-label for each type of motif to write about in the paper.
- MotifA - Lenalidomide <–> Thalidomide –> TNF [Multiple Myeloma among other things]
- MotifB - Topiramate
- MotifC - Isradipine [Heart/Blood Pressure and Parkinson's Disease/ ~Paralysis]
Ideas for the scoring section of the paper:
- Check the values such as Tanimoto, BLAST and G-Sesame known motifs (i.e. Chlorpromazine)
- Is there any differentation between a known motif and a random unsignificant motif
Other Tools
Jul 7th
Created txt files of the network, removing the semantics:
- 1 File seperated by spaces
- 1 File separataed by tabs
In the format:
node# relation_type node#
Papers
Jul 5th
Some papers
- Itzkovic, S. and Alon, U. (2005) Subgraphs and network motifs in geometric networks. Physical Review. 71(2)
- Wernicke, S. (2006) Efficient Detection of Network Motifs. Transactions On Computational Biology and Bioinformatics. 3(4):347-359
http://www.slideshare.net/florianmarkowetz/network-biology-lent-2010-lecture-1-3322887
DOT Export appears to work but the file output needs the semantics removed (e.g. node24 [label="Compound 24", shape=box, color=white, width=1.63, height=1.63, fixedsize="true"]).
Stats
Jul 1st
- 4842 – Drugs
- 3500 – Targets
- 10743 – bi_to
- 274607 – h_s_s
With this information how likely/often will a motif appear in the dataset.
ONDEXView hss = graph.getRelationsOfRelationType(graph.getMetaData().getRelationType("h_s_s"));
int j = 0;
for (ONDEXRelation r : hss) {
ONDEXConcept to = r.getToConcept();
ONDEXConcept from = r.getFromConcept();
if (to.getOfType().getId().equals("Protein") && from.getOfType().getId().equals("Protein")) {
j++;
}
}
System.out.println("hss " + j);
Improvements
Jul 1st
TypeA
- Removed all motifs where both similar drugs bind to the target.
- 30,231 motifs were originally identified, now 27,499 (-2,732).
TypeB
- Removed motifs where the drug binds to both targets.
- 22,864 motifs were orignally identified, now 19,185 (-3,679).
TypeC
- Removed motifs where protein(1) inv_in disease(2) or protein(2) inv_in disease(1).
- 70,866 motifs were orignally identified in the dataset, now 70,715 (-151).
- Tried: Remove motifs where the target connects to protein(2), this did not take into account that some links are has_similar_sequence/structure and not is_a. 200 (-70,515).
- Removed motifs with target is_a protein2, made no difference to total.
Ondex
- Selecting a node with a certain id in Ondex?
- Too hard when there is 500k nodes
- Search doesn’t work
- Ondex Console: how to use this??? -> help is unavailable surprisingly.
