Newbie: applying a set of values to an arbitrary column

KNIME Analytics Platform
1

I’m working on a bioinformatics pipeline (PCR data) to demonstrate knime’s capabilities to researchers. However, as a newbie to knime I’m stuck :wink:

What I’d like to do is apply a set of numbers:
“Gene”,“Min(Mean(Ct Value))”
“HKG1”,24.65
"HKG2",18.35
"HKG3",32.349999999999994

To an existing table:
“Gene”,“Sample”,“Min(Ct Value)”,“Mean(Ct Value)”,“Standard deviation(Ct Value)”,“Unique count(Ct Value)”,“First value(Amplification Efficiency)”
“HKG1”,“sample1”,24.5,24.65,0.21213203435613576,2,2
"HKG1",“sample2”,25.4,25.6,0.28284271247408843,2,2
"HKG2",“sample1”,18.3,18.35,0.0707106781186226,2,2
"HKG2",“sample2”,18.7,18.75,0.07071067811942648,2,2
"HKG3",“sample1”,32.3,32.349999999999994,0.07071067812264203,2,2
"HKG3",“sample2”,33.0,33.25,0.3535533905932738,2,2

with no additional grouping in the second table. So, for example, a new column called “Min(Mean(Ct Value))” would be created in the second table with 24.65 inserted for every gene with HKG1 in the first column. Ditto for all the other rows in the set of numbers. Although I could hardcode it with something like the rule engine, I want it to be adaptable to different datasets.

But seemingly no combination of concatenate, joiner, and groupby nodes seems to do what I need. I think I’m missing something, but cant see what…

2

Hi,

Because the Joiner uses the RowId from the first first table input your first dataset (the reference table in this case) as the first input and then join the dataset as your second input on the field “gene”.

This will give you what you are looking for. If it doesn’t work could you please provide a little more detail on the data itself.

Thanks,

Jay

3

Hi Jay,

Woohoo! Thanks for that! I was making the mistake of not putting the Gene column in the reference table into the row ID (just leaving it as a column). I couldn’t figure out why the join was blank…

4