NEWS.md

In DeLorean: Estimates Pseudotimes for Single Cell Expression Data

Version 1.5.0

• Use pre-compiled models as in rstantools.
• Rename models to remove hyphens. Backwards compatibility is conserved however.
• Do not expose Stan functions as this is impossible when using rstantools framework.

Version 1.4.2

• Use updated version of roxygen2.

Version 1.4.0

• Make changes to how compiled models are cached to satisfy silly CRAN policy. Now compilation happens more often: once per model per R session.

Version 1.3.0

• Remove filter.genes and filter.cells, (renamed to use underscores) in order to pass CRAN checks.

Version 1.2.6

• Deprecate filter.genes and filter.cells, (renamed to use underscores).

Version 1.2.5

• Update rstan requirement to 2.14.1 to avoid bug: http://andrewgelman.com/2017/01/01/stan-2-14-released-r-python-fixes-bug-sampler/

Version 1.2.4

• Increase border of inducing points in sparse approximations
• Fix further Stan bug (feature?) with square root of integers
• Update rstan requirement to 2.12.1 from bad 2.10 version: http://andrewgelman.com/2016/07/31/stan-2-11-good-stan-2-10-bad/

Version 1.2.3

• Update for new package versions: rstan (2.10.1) and dplyr (0.5.0).

Version 1.2.2

• Updated code for revision of manuscript. Changed cell size normalisation strategy.

Version 1.2.1

• Fix for sqrt(integer) bug on Solaris x86.

Version 1.2.0

• First version submitted to CRAN.

DeLorean documentation built on May 2, 2019, 9:24 a.m.