VPSPulse Mirrors
High-Performance Open-Source Archive
NEWS
ampir 1.1.0 (29/06/2021)
- ampir now has a Shiny based web interface! https://ampir.marine-omics.net/
- updated both classification models (ampir_precursor and
ampir_mature) with additional AMPs
- fixed an example in the README which resulted in NAs output (@Ebedthan, #20)
- converted input for
calculate_features.R to a
dataframe, in case it was not a dataframe
ampir 1.0.0 (11/05/2020)
This
is new release which contain new features and bug fixes. In this
version, we have:
- added the
remove_non_standardaa.R function
- exported
calculate_features.R
- tidied up code to improve running speed of
calculate_features.R
- changed the default min_len parameter in
calculate_features from 20 to 10
- included the option to parallelise
predict_amps.R
- inluded the option to add self-trained models to
predict_amps.R
- added another vignette (
train_model.Rmd) detailing how
to train models for predict_amps.R
- updated the default prediction model used in
predict_amps.R and optimised it for precursor proteins
- added a secondary prediction model for use in
predict_amps.R, optimised for mature proteins
- updated introduction to ampir vignette (
ampir.Rmd) and
README.Rmd
ampir 0.1.0 (20/11/2019)
Initial release to GitHub
- Prior to this, it was a private package
Initial submission