Error when NB model doesn't fit: while ((it <- it + 1) < limit && abs(del) > eps) missing value where TRUE/FALSE needed

[karinkumar]

Hello @walaj and I were trying to run a fishhook on sparse data (very few breakpoints across the binned genome). Sometimes the data is sparse enough that the nb model doesn't fit. In fact, it throws an error that's within glm fitting function : Error in while ((it <- it + 1) < limit && abs(del) > eps) { : missing value where TRUE/FALSE needed

We are switching to a poisson model, but perhaps there is a better way to catch this error and print a message or redirect to a poisson model instead. Right now it just stops the fishhook model from running.

[mskilab]

Totally agree - I think that's a good suggestion. We will implement

…

On Thu, Nov 1, 2018 at 10:48 AM Kiran Kumar ***@***.***> wrote: Hello @walaj <https://github.com/walaj> and I were trying to run a fishhook on sparse data (very few breakpoints across the binned genome). Sometimes the data is sparse enough that the nb model doesn't fit. In fact, it throws an error that's within glm fitting function : *Error in while ((it <- it + 1) < limit && abs(del) > eps) { : missing value where TRUE/FALSE needed* We are switching to a poisson model, but perhaps there is a better way to catch this error and print a message or redirect to a poisson model instead. Right now it just stops the fishhook model from running. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#3>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AL-0keLvmgjq_LAav6DSi6hHvWXxD9CCks5uqwmvgaJpZM4YGgjg> .

[karinkumar]

Thanks!!