Stijn van Dongen

This sounds great Paolo. And this: ``` It should even be possible to pipe the result of event handles to a channel, e.g. foo.onError { /* logic */ } |...

Hello, there are quite a few ideas on how to approach this. How would you feel @pditommaso about the following small new feature: - an output channel can declare e.g....

Hi, this is a known error - apologies. please see my comment+solution in [this discussion](https://github.com/micans/mcl/discussions/25#discussioncomment-7939818). I'll keep this issue open until the next release so that people can find the...

Hello, could you perhaps read the exchange of all comments in #8; in that case a graph of size 31965032x31965032 with 107744198225 nodes was clustered. Your command was being killed...

Hi, I don't have experience with metagenomic data. I think mcl may have been used with it, I remember seeing publications that used it, but this may have been on...

Thank you for your kind words. I am a little curious why you have 12T of data. The network in #8 has more nodes and edges and was reported as...

I suggest to only use binary format, so the first version. To test you can simply do ``` head -n 1000000 data | mcxload -abc - --stream-mirror --write-binary -o data.mcx...

I see two aspects here. First the paper states: *To identify genus- and family-level vOTUs, we clustered viral genomes using a combination of gene sharing and AAI. For computational efficiency,...

Am I correct in that you already loaded this data before and saved it in `data.mci`? In that case you can just do ``` mcx convert data.mci data.mcx ``` After...

- `mcxload` is the only way, correct. Loading billions of edges will always take a lot of time. - there are ways to use multiple machines but this requires a...