target

[Instructions]
target
Search for CNV alterations. Parameters for the query are:
Target1 : the HUGO symbol of the gene you are searching for
Target2 : the HUGO symbol of the second gene for either co-occurance or breakpoint

co-occur will search for samples which contain a segment that meets the criteria for both Target1 AND Target2

breakpoint will search for a segment break within the start/end of both Target1 and Target2. This type of query can be used to evaluate potential unbalanced chromosomal translocations for possible fusion genes. However, the multitude of segment breaks makes this more suitable for verification rather than discovery. Additionally, balanced translocations maybe undetectable by the segmentation algorithm due to smoothing. Visual inspection maybe necessary.

Sort : determines the order the CNV segments are listed.
mean - by the CNV segment value (default)
cell - alphabetically by the sample name
expression - if an expression dataset is selected, sort by the expression value
mutation - if a mutation dataset is selected, sort by the mutation field

Breakpoint pad : if performing a breakpoint search, add the number of bases to the start/end coordinates for each gene to allow for finding breaks near the coding region that don't actually break the coding region. (eg. promoter fusions)

min : the minimum segment size required in kilobases. (default 100KB) Sizes smaller than 10KB may return segments with spurious probes and germline SNPs. However, certain micro-deletions may only be visible this way. Use 0KB to return all samples.

max : the maximum segment size allowed in megabases. (default 10MB) Use this parameter to restrict results to smaller segments vs. large arm level chromosomal events. Because large segments may contain hundreds of genes, the role of any one gene in oncogenesis maybe more uncertain.

mean : restrict segments by the mean log2R. (default 2) Remember that the segments are ratioed to an average of HapMap samples which are diploid. Thus, a value of 0 is equivalent to 2^0=1 diploid genomes or 2 copies. A value of 1 is 2^1=2 diploid genomes or 4 copies. For deletions, use negative values. Generally the range of CNV is -2 to 2. For either positive or negative values use absN. eg. abs1 for >= 1 or <= -1

Images : show CNV plots inline, instead of rollover. (default unchecked) Use this option if viewing CELLX on an iPad because the rollover is difficult to perform on a touch device.

Additional parameters:

CNV SOURCE : the CNV dataset to search. (required) ALL_HG18 will search the whole database of SNP6.0 chips. ALL_HG19 will search all the Illumina chips.

EXPRESSION : if the CNV dataset contains the same sample names as in the expression dataset, merge the expression data into the table. Will still try to merge anyway even if samples do not match. Useful to mix and match cell line CNV/Expression data. (optional)
AFFY HUGO : display the expression of a different gene than the one queried for. (optional, requires EXPRESSION)
MUTATION : show mutations by merging on the sample name. (optional)
MUT HUGO : display the mutations of a different gene than the one queried. (optional, requires MUTATION)

Tissues : restrict results to the selected tissue type, if sample is annotated. (optional)
Cell Type : add columns to the results table listing tissue type. (default unchecked)
CNV/EXP plot : simple scatter plot of CNV mean vs. expression columns. (default unchecked) deprecated, use [expression_plot] instead. The plot can be misleading if the same sample is listed multiple times due to multi-segments within the same gene.

Cell Lines : restrict results to explicitly listed sample names. (optional) (may not work for TCGA samples due to the naming scheme)

Example: click for example query