- minor change, for another package to be able to access one of the
internal functions, it was exported
- better checking of dataset inputs, and better documentation of what
that data should look like
- update to new susieR >= 0.12.06 which requires sample size
- general tidy up of susie-related code
- new functions to help find subregions for finemapping - findpeaks,
- updated FAQ.md
- tidied empty items in documentation to make CRAN happy
- slimmed down example dataset to make CRAN even happier
- deprecate runsusie() argument p
- update plot_dataset() and sensitivity() to fix bug for very small p
values displaying as 0
- release to CRAN of coloc.susie
- remove nref now not recommended
- in runsusie, allow nref to be overridden if user passes
- deprecate coloc.signals for multiple causal variants
- introduce coloc.susie for multiple causal variants
- new vignette 02_data giving more details on structuring your data
- missing sdY for type=“quant” would error in finemap.signals or
coloc.signals, now it is estimated if missing.
- warning for factors used as snp names added (issue #29)
- snps in the second dataset might not have been masked as intended.
This would result in effect using the “single” option for trait 2 when
masking was expected.
new functions: coloc.signals, finemap.abf - analogues of
coloc.abf and finemap.abf that allow for multiple causal variants. See
vignette on conditioning/masking new function: sensitivity -
post-hoc, determine the sensitivity of coloc results to changes in the
prior. See vignette on sensitivity
- BUGFIX: finemap.abf()
- in low power situations, the posterior for H0 was previously too
low. This will only affect datasets where the minimum p value was >
1e-7 - ie where the posterior for H0 would be expected to be much above
- fix bug in process.datasets which suggested MAF was needed for cc
data when beta/varbeta also present
- added pkgdown
new function: finemap.abf - improved clarity of error
messages in coloc.abf() sub functions - added finemap.abf to fine map a
single trait - fixed warnings caused by CRAN disliking the BioConductor
- added stratification to proportional testing approaches
- tidied code relating mainly to proportional
- colocalisation testing methods, making more methods confirm to
- pcs.prepare now imputes missing genotypes by default 2013-05-22
- Introduced a function to estimate trait variance from supplied
coefficients and standard errors. This is used within the approach
implemented in coloc.abf(), and replaces the earlier version which
implicitly assumed that var(Y)=1 for quantitative traits, which could
lead to incorrect inference when var(Y) was far from 1.
- Merged coloc.abf and coloc.abf.imputed(), so that datasets for which
beta, var(beta) are available can be matched to datasets with only p
values and maf.2 This means the arguments to coloc.abf() have been
changed! Please check ?coloc.abf for the new function.
- Bug fix for coloc.abf() function, which used p12 instead of log(p12)
to calculate L4.
- New function coloc.abf.imputed() to make better use of fuller
information on imputed data.
- New function, coloc.abf(), to implement the colocalisation approach
described by Giambartolomei et. al.
- Changes in the coloc.test() and coloc.bma() functions to make them
consistent with regards arguments and output.
- added principal components functions pcs.prepare(),
- Restructed the coloc objects to separate Bayesian and non-Bayesian
- added Credible Interval calculation to coloc.test().
- updated to return u and Var(u) in addition to chisq statistic.
- fixed error in documentation, added MASS to Depends.
- made the means of generating plots more flexible.