"Target genome is too different from training genomes" with every dataset used

[Elcquid]

Hello, I have tried datasets of up to a thousand genomes, but no matter what I do, I get failure to converge. Is there something I'm doing wrong? Here is an example of how I'm running the code:

./fitModel.py -t GCF_000218655.1_Fusobacterium_sp_11_3_2_V1_genomic.fna -sp 'Fusobacterium nucleatum' -s 1000

Thanks

[jccastrog]

Hello Johnny,

Thank you for your patience. I can try to replicate the error and get back to you.

A couple things could be affecting the result off the top of my head. First the version of the scipy.optimize you are running is somehow different from the one I use so it might lead to low (or in this case) no convergence. Or it could be that the genomes for your species are way too close genetically to make a meaningful distinction. In which case there are a couple of options we can think of, but it depends on the biology of the bug.

Let me know if this helps and I'll try to get back to you early next week

Juan

On Wed, Jul 12, 2023 at 2:24 PM Johnny Connolly @.***> wrote:

Hello, I have tried datasets of up to a thousand genomes, but no matter what I do, I get failure to converge. Is there something I'm doing wrong? Here is an example of how I'm running the code:

./fitModel.py -t GCF_000218655.1_Fusobacterium_sp_11_3_2_V1_genomic.fna -sp 'Fusobacterium nucleatum' -s 1000

Thanks

— Reply to this email directly, view it on GitHub https://github.com/jccastrog/imGLAD/issues/2, or unsubscribe https://github.com/notifications/unsubscribe-auth/AC3JEH22KHA6KRCKVVGBCDLXP3TVXANCNFSM6AAAAAA2H3WJF4 . You are receiving this because you are subscribed to this thread.Message ID: @.***>

[Elcquid]

Juan, Sorry about the delay, I was away from work, and you responded faster than I expected! The scipy we're using is 0.18.1. Which should I be using?

Thanks, Johnny