- What is a molecular target?
- How do I locate a target of interest to me?
- Can I see if a target is correlated with sensitivity or resistance to any drugs?
- Why are there multiple entries for a given gene?
- How can I find out more about a Molecular Target?
- My Molecular Target correlates with a compound. Does that mean the compound is directly inhibiting that Target?
- I would like to measure a target. How do I go about it?
- Can I download the data?
- How can I link my website to the Molecular Targets?
- Useful links
Researchers from NCI as well as numerous extramural researchers have obtained samples from the NCI's panel of 60 human tumor cell lines and measured a variety of targets in these cell lines. Targets include gene mutations, mRNA levels, protein levels and enzyme activities.
Many targets can be linked to one gene (e.g. mutation of gene X, mRNA levels of gene Y, protein levels of gene Z) or to several genes (e.g. enzyme activity of a family of genes). If you know the HUGO approved name for a gene you can type that directly into the search by "HUGO approved gene name" query box. A great search site for HUGO approved names can be found at the Weizmann Institute's GeneCards site. Just type a key word (e.g. cyclin, p16, kinase, an accession number, a Unigene number) for your target in the query box, and a list of possible genes will result.
Many genes are known by multiple names. To search for a target by a name other than the HUGO approved gene name, type that name (or a portion of the name) into the search by "Alias" box.
Alternatively, gene-related targets can be browsed from the alphabetical menus.
Some targets cannot be linked to a gene. These targets can be browsed from the "All targets" menu.
For some Targets, more than one researcher has measured the Target. For some Targets there are several sorts of measurements (e.g. protein and RNA).
Some Targets have more than one pattern associated with a single experiment. For example, for some data we have done a logarithmic transformation of the data, generating a new pattern that can be used as a seed for a COMPARE search. The Experiment ID # identifies a single experiment. A Pattern ID # identifies a single pattern. A Experiment ID can have multiple Pattern ID's associated with it (e.g. a primary pattern and a log-transformed pattern), which are just different ways of analyzing the same underlying data.
Yes! Currently you can begin with a compound of interest and run COMPARE against either the individually measured Molecular Targets or against the Brown/Botstein/Weinstein microarray dataset.
Under development: Once you click on a target name you can choose to run COMPARE. This program compares the patterns of drug sensitivity in the NCI panel of 60 human tumor cells for about 60,000 compounds to the pattern for your chosen Molecular Target (the seed). Results will be given as a correlation coefficient ranging from 0 to 1.00. Choose STANDARD COMPARE to identify compounds or Targets that correlate with high levels of the Molecular Target. Choose REVERSE COMPARE to identify compounds or Targets that correlate with low levels of the Molecular Target. To find compounds for which high levels of a Molecular Target confer resistance, choose REVERSE COMPARE. The closer the value is to 1, the better the correlation.
Once you click on a target name you will see links to various sites with information about that target.
- The publication where the Target measurement is described.
- Protocols used to measure the Target.
- The home page of the laboratory that measured the target.
For gene-based targets there are additional links:
- GeneCards at the Weizman Institute provides information on function, disease associations, catalytic activities and sequence resources.
- Unigene at NCBI provides DNA sequence resources.
- Gene at NCBI provides information on function and sequence resources.
- OMIM at NCBI provides information on disease associations.
No! These correlations should be looked at as hypothesis generators. The compound might inhibit the Target. Or, perhaps the compound interacts with a related target (maybe a component of the same metabolic pathway). The compound might cause some sort or damage that is repaired by the Target. The Target might regulate genes whose products are inhibited by the compound. The compound might interact with a cellular component and trigger a checkpoint that requires the Target. Etc.
Any correlation of a Molecular Target and a compound should be verified by designing experiments to test such hypotheses.
Fill out the Molecular Target Proposal form and email it to:
Yes! See our Molecular Target data download page.
Yes. Use the following format to create a url (link) to a page for a particular molecular target:
XXXXXX is the molecular target identifier
ZZZZZZ is the experiment number
- Microarray data and analysis from the Brown and Botstein group at Stanford University.
- Microarray data and analysis from the Weinstein group at the National Cancer Institute.
- The IARC Database of somatic p53 mutations in human tumors and cell lines
Email questions concerning DTP's molecular targets program to: Susan Holbeck [firstname.lastname@example.org]