Many to one join

Hi,

So this is probably a long shot but...

I am trying to create a user-friendly, open-access compound data curation protocol. As part of this, my data set needs to be filtered for certain elements/functional groups as some curation/transformation is not suitable for all compounds.

Now, my current issue is that when I have separated some sets of compounds into different curation pathways, when I try to bring them together, there are no nodes (that I can find or are obvious) that will allow me to link 3 or more pathways together. They all have at least 4 columns of data similar in each pathway so finding a common column header (ie. compound_id) is not an issue.

I can only find 'Joiner' which doesn't work for this as it looks for matching data and is only a 2-1 join.

Does anyone know of anything that can bring disparate data paths together?

thanks

Isn't the concatenate optional in node what you want.

also the End case node will work for you too.

simon.

Hi Simon

No, unfortunately the concatenate can only allow 2 paths to combine. The End Case only allows 3. All I want, and this is common in Pipeline Pilot, is to just feed in multiple paths (4+) into the same node. No manipulations required, just combining data as they will all have common columns/data rows.

The end case won't work as there are some extra columns coming from some pathways that aren't in others. In PP, this is just added as an extra column for all data with the rows being populated if present. 

thanks

Louisa

Louisa,

As Simon said what you need is the "Concatenate (Optional in)" node.  Should do what you need - it has four input ports.

Alastair

Alastair

That only has 4 input ports, but I can see that i could just use a couple. There seems to be some lost data when I've joined multiple ports but that might be resolved by used a molecule->type cast node.

Louisa

Yes, often I have huge chains of multiple Concatenate OPtional In nodes, although I dont recall ever seeing any data loss.

Alastair