Filter Agilent Data

Question

[j_jamal96]

Hi everyone,

Could you please help me to know the answer to below question: What does this part of your Agilent data show ''AgilentData$other$gIsWellAboveBG"

My data has three different Cell-lines, each containing a resistant group and a sensitive group with two repetitions. My goal is compare the sensitive and resistant groups, I don't want to do the contrast between cell lines. Despite this fact, what number should I use for filtering

(Filt_Data <- rowSums (AgilentData $ other $ gIsWellAboveBG> 0)> = **????**)       *2, 3, ... ????*

If I want to do the contrast, would be different with this above mentioned number?

Thanks 💐

Answer 1

[Kevin Blighe]

Hi,

The decision of which number to use is the analyst's choice. If your dataset has 16 samples, then you may choose a value of 8, so that only probes that have signal (above background noise) in >=50% (8 / 16 =~ 50%) samples are retained.

Kevin

[j_jamal96]

NO, 12 samples ...

Cell line 1 Resistance 2 parebtal 2

Cell line 2 Resistance 2 parebtal 2

Cell line 3 Resistance 2 parebtal 2

[Kevin Blighe]

12 samples - great. As the analyst, you can make the decision about which threshold to use.

[j_jamal96]

I want to compare the resistant and sensitive groups of each cell-line, but I don't make the contrast between the cell-lines. So our analysis is divided into three separate sections (three cell-lines).

The AgilentData$other$gIsWellAboveBG matrix has 12 columns, that four of which related to a cell-line, respectively. so that only probes that have signal in >=50% (In each of the cell-lines) I used the following code for the filter ...

EProbe <- rowSums(y$other$gIsWellAboveBG[ , 1:4] > 0) >= 2 | rowSums(y$other$gIsWellAboveBG[ , 5:8] > 0) >= 2 | rowSums(y$other$gIsWellAboveBG[ , 9:12] > 0) >= 2

what is your idea?

[Kevin Blighe]

I think that you may want to consider the logical operator of use:

• | == OR
• & == AND

Other than that, I do not see any problem.