Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

lucas

About

Username
lucas
Joined
Visits
271
Last Active
Roles
Administrator

Comments

  • Glad I could help :-) Lucas
  • Hi Alicia, okay, it seems I should go a bit more into detail ;-) You said "a branch with the highest LWR might be formed by tons of placements with 0.00001 LWR". Well, let's assume "a ton" of placements are 1000. Then, this branch would only get a…
  • Hi Alicia, the demo visualizes the distribution of the likelihood weight ratios, as this is the more "detailed" view on the data. If by "the first part of the code" you mean this function: https://github.com/lczech/genesis/blob/master/doc/demos/vis…
  • Hi Charles, there was yet another small issue with the Earth Mover's Distance values in the "Compare Jplace Files" demo program. If you are using those values, please update to Genesis v0.12.1. Sorry for the inconvenience, I really should check m…
  • Also, if you are still needing the "Compare Jplace Files" demo program, please update to genesis v0.11.1: https://github.com/lczech/genesis/releases/tag/v0.11.1 I recently found a bug in the demo that can lead to some wrong entries in the second ta…
  • Ah okay, thanks, got the file. The issue was the following: I did not pay too much attention to efficiency in the placement_classification_table app, so for each placement, a distance measure on the whole tree was calculated. That was easy to fix.…
  • Hm okay, no idea what's going on there. Can you please email one of those files to me?
  • Hi Charles, don't worry, ask as many questions as you need until stuff is working. Otherwise, my programs would be pretty useless, wouldn't they?! Well, from the distance, it is hard to judge what is going on there. How big is the jplace file, par…
  • Hehe nice. I was already wondering what was going on there... Probably just some old files that were not properly updated.
  • Hi Charles, wow, that's quite some errors. Can you try `make clean`, then `make` in the v0.11.0 version, and post the whole output again? Thanks Lucas
  • Hi Doug, I just released genesis v0.10.0, which also contains a demo program for comparing two jplace files: http://doc.genesis-lib.org/demos_compare_jplace_files.html It should cover your case. However, you probably don't need to worry about the …
  • Hi Doug, that's interesting. How long are the query sequences you are using? If they are short (
  • Hi Doug, I just spoke with the mPTP guys in our group. It is expected that it gives you fewer counts than GMYC and PTP. mPTP penalizes if there are too many parameters, i.e., too many species - thus it yields more robust, but lower numbers. If yo…
  • Hi Charles and Doug, we recently released a significantly faster and more accurate version of our species delimitation tool PTP: Preprint: http://biorxiv.org/content/early/2016/07/14/063875 Web-Service: http://mptp.h-its.org/ Documentation & D…
  • Hi Charles and Doug, we recently released a significantly faster and more accurate version of our species delimitation tool PTP: Preprint: http://biorxiv.org/content/early/2016/07/14/063875 Web-Service: http://mptp.h-its.org/ Documentation & D…
  • Hi Charles, well, speaking from a computer science perspective, if your data tells you that you have that many OTUs, then you should not try to artificially enforce fewer. In particular, SWARM gives you "natural" OTUs. That means, it clusters all r…
  • Hi Charles, your idea sounds interesting. However, evolutionary placement was not developed as a method for OTU clustering, so it doesn't surprise me that you got unsatisfying results. Usually, you'd want to use some OTU clustering method first, an…
  • Hi Charles, yes, if you do not have that many query sequences, I'd expect oversplitting if you use EPA as a form of OTU clustering. We are actually currently working on a comparison of exactly this approach (including a downstream analysis with PTP…
  • Hi Charles, all right, then here is your chance to learn some C++ ;-) You find the code here: https://gist.github.com/lczech/24c12a912e4eb0ed38d517d887c42db5 Works as always, put it in the "apps" directory and call "make update". The program expe…
  • Hi Doug and Charles, I just released genesis v0.8.0 (https://github.com/lczech/genesis/releases/tag/v0.8.0), where instead of this complicated clade format with different files in a directory, we now use a single file with tab separated lines for e…
  • Hi Charles, that is a simple task. Maybe you want to try it on your own - would be a good exercise to start working with the genesis API ;-) A good starting point is the placement tutorial: http://doc.genesis-lib.org/placement.html There, the infor…
  • Hi Charles, hi Doug, this is not particularly difficult. I wrote a program which extracts names of pqueries for each of the edges of the reference tree: https://gist.github.com/lczech/f42ed5c6f229a078449b308ad5497919 You need to update to genesis v…
  • Hi Carlijn, that was an unfortunate combination of g++ and Boost Python which is not (yet) covered in the automated tests. I resolved it now. Please get the updated genesis v0.6.0 (https://github.com/lczech/genesis/releases/tag/v0.6.0) and let me k…
  • Note to the note: added automated testing of the demo programs before releasing a new version. ;-) Goodbye compilation errors.
  • Hi again, also, if you know of any file format to store clades, let me know. Would be a nice feature to read those instead. Lucas
  • Fixed it, see genesis v0.5.1: https://github.com/lczech/genesis/releases/tag/v0.5.1 Note to myself: test all demo files before releasing a new version ;-)
  • Remark: Extracting the clades from a tree is a bit cumbersome to do manually, I know. Unfortunately, it is inherently a manual task, as it is your choice which subtrees you want to group into a clade. Tree visualization tools (like Dendroscope) can …
  • Hi Charles, I just released a new version of genesis (https://github.com/lczech/genesis/releases/tag/v0.5.0), which contains example files for the demo. You find instructions in the manual (http://doc.genesis-lib.org/demos_extract_clade_placements.…
  • Hi Doug, thanks for your feedback, and sorry for the late answer, I was sick the last couple of days. I just released a new version of genesis (https://github.com/lczech/genesis/releases/tag/v0.5.0), which contains, example files for the demo. You…