Scanpy Parameter Iterator issues

[wee-snufkin]

I tested Scanpy Parameter Iterator tool and built a tutorial based on it. Here are my comments what could be improved:

• From the user point of view, it would be very useful to have tags with the chosen values automatically added to each dataset in the output collection. Currently the values are added to the name of the parameter such as "perplexity_1", "perplexity_5" but for downstream analysis it's not as useful as tags would be.
• "Inputs" section in tool description doesn't specify that n-neighbours (for Scanpy ComputeGraph) can also be an input (takes only integers)
• ‘Step increase values to be iterated’ returns only floats, gives an error for integer-requiring parameters
• I tested perplexity with floats and it worked well so I suppose that in fact it CAN take floats
• Would be good if the used parameter values were added to the individual dataset names in Galaxy history so that it's easier to track and wouldn't require using tags
• Is there any way we could use Parameter Iterator on the collection output of a tool used with another Parameter Iterator? Example: we use ComputeGraph with Parameter Iterator and get a collection with 5 datasets. If we passed that collection to RunTSNE tool and provide 8 parameter values with another Parameter Iterator, we would expect 40 output files (each of 5 datasets run with 8 different values), while in reality, we get only 5 as the collections match their content one-by-one, ie knn_1 + param_1; knn_2 + param_2 etc instead knn_1 + param_1; knn_1 + param_2…
• I encounter a weird problem when testing the tool and I described it in the tutorial as follows: "In Use programme defaults you can specify Additional suffix to the name of the slot to save the embedding – if it’s included, PERPLEXITY will be substituted with the value of the perplexity setting. However, in that case you will get an error when using Scanpy PlotEmbed: due to the value included in the entry name, the tool will not recognise the correct embedding. Therefore, we leave that box unfilled even though it would make it easier to differentiate between datasets."

Overall great tool, saves much time - thanks!