RNAi Screens for Genes Involved in Golgi Glycosylation - Membrane Trafficking - page 422

5. Now determine a primary hit list of genes: for each lectin,
combine data from all plates and plot lectin intensity per cell
against cell count ( see Notes 14 and 15 ).
If there is a dependent relationship between the two parameters, as
occurs for many lectins:
(a) Fit a curve to the plot and obtain the fi tted equation
( see Note 16 ).
(b) For each data point (gene), calculate the absolute devia-
tion ( d i , see Fig. 3 ) of the experimental y value from the
fi tted y value (determined by fi t equation) for the same cell
count ( x i ). Then determine the median absolute deviation
among all genes ( see Note 17 ).
(c) Decide on the cutoff value to defi ne hit genes, such as fi t-
ted y value ± k median absolute deviation. A gene is then
considered a hit if it has a value greater than fi tted y
value + k median absolute deviation or smaller than fi tted y
value − k median absolute deviation ( see Note 18 ).
If there is no dependent relationship between the two parameters:
(a) Determine the median and the median absolute deviation
of all genes ( see Note 17 ).
(b) Decide on the cutoff value to defi ne hit genes, such as
median ± k median absolute deviation. A gene is then con-
sidered a hit if it has a value greater than median + k median
absolute deviation or smaller than median − k median
absolute deviation ( see Note 18 ).

1. Before beginning, test the double-knockdown protocol on a
small scale and optimize if necessary. As a starting point, use an
equal amount of glycosylation enzyme-targeting siRNA as the
amount of hit gene siRNA and the same amount of HiPerFect
transfection reagent as in the primary screen ( see Note 19 ).
2. Thaw plates containing siRNAs targeting candidate hit genes in
the cell culture hood for 30 min. Use duplicates of each plate.

3.4 Secondary
Assay: Double
Knockdown for Lectin
Signal Validation

3. See step 2 of Subheading 3.2 .
4. See step 3 of Subheading 3.2 .
5. In one replicate, dilute the siRNA targeting the relevant glyco-
sylation enzyme into Opti-MEM such that the amount of
siRNA dispensed is equal to that of the candidate hit gene
siRNA in the well. The volume dispensed here should be about
half that in step 4 of Subheading 3.2 . In the other replicate,
use non-targeting siRNA instead.
6. See step 4 of Subheading 3.2 . The volume dispensed here,
together with that in the previous step, should add up to the
volume in step 4 of Subheading 3.2 .
7. See steps 5 - 8 of Subheading 3.2 .