The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.
GEOquery package, as it may no longer be necessary for
preparing GEO series data.plot*
functions will be enhanced to allow greater user customization.dgeList, which will offer improved customization
and performance.list2dt() and list2graph() are the
exception and keep their original names: the 2-for-“to”
idiom matches base R (list2DF(), list2env())
and both names satisfy the snake_case linter, so renaming them would
have broken a released name for no benefit. The briefly considered
list_to_dt()/list_to_graph() never reached a
release.avg_log2fc_threshold, top_cell_n,
tissue_class, tissue_type,
min_count, min_unique_n,
ignore_case, group_column,
sample_info, feature_info,
data_text, gsea_param,
ticks_size, fgsea_res).prepare_tcga() returns expr_count and
expr_fpkm instead of exprCount and
exprFpkm, so every field of its two tables is snake_case.
The old names still work and warn; they will stop working in 1.4.0.data-raw/cellmarker.R).tuneParameters() now stores the annotation in the
meta.data column “CellMarker3.0” instead of “CellMarker2.0”.matchCellMarker2() renamed to match_ref()
because it also supports custom reference datasets. The old name is
deprecated and will be removed in the next version.cellMarker2 to cellMarker3 (internal; reflects
the CellMarker 3.0 source).match_ref() ranks candidate cell types by
uniqueN (breadth of matched markers) with N as
tie-breaker instead of by N alone; a single heavily
reported marker could otherwise outweigh a cell type matching on dozens
of markers.match_ref() reports the detection rate
(pct.1) of each matching marker in the new
pct_with column, aligned with ordered_symbol
and NA when the input has no pct.1. It is
reported for auditing and does not affect the ranking; see the
min_pct argument below if you want detection rate to
influence the annotation.match_ref() gains a min_pct argument:
markers whose detection rate in the cluster they were found for
(pct.1) is below it are dropped before matching, so an
annotation cannot rest on genes that are barely detected. It defaults to
NULL (no filtering, i.e. the previous behaviour) and is
ignored, with a message, if the input has no pct.1 column.
Note that n is applied after this gate, so the markers used
are not a subset of the unfiltered ones.
Seurat::FindAllMarkers(min.pct = ) cannot replace it: that
argument gates which genes are tested in either population, so a gene
detected in a handful of cells can still end up as a positive
marker.plot_possible_cell() gains a value
argument: "N" (the default, unchanged),
"uniqueN" (the measure match_ref() ranks
candidates by) or "pct", which draws the share of matching
markers that are actually detected at min_pct as tiles
labelled with uniqueN — the evidence check for an
annotation. value = "pct" errors when the input carries no
detection rate. min_unique_n now includes candidates
matching exactly that many markers; it previously required more than
that.available_tissue_type() gains a
tissue_class argument, so the tissue types recorded under
one class can be listed
(available_tissue_type("Human", tissue_class = "Blood"))
before both are passed to match_ref(). A class the species
does not have is an error rather than an empty result:
tissue_class and tissue_type are two labels
recorded per database entry, not a hierarchy, so whether a pair exists
is worth checking rather than assuming.match_ref() warns when the tissue filters leave an
empty reference set, instead of silently returning a table with no
candidate in it. A class and a type that never occur together on a
database entry select nothing, which is easy to write by accident
because the two fields look like a hierarchy;
available_tissue_type() above is how to check.tissue_class = c("Blood", "Bone marrow"), with the
reasoning: the default searches every organ, and bounding the class is
what makes the candidate list readable, while bounding
tissue_type as well removes evidence rather than correcting
for anything.R CMD check, CI and Rscript stay
quiet.Seurat::DotPlot in vignette to avoid upstream
upgrade’s affectscheck_marker, plotSeuratDot).matchCellMarker2).finsert.get_marker).data.table version to 1.15.0plotSeuratDot.tissueType and
tissueClass in single cell related function.Artist Class: Now includes a default
data argument for data exploration. All commands and results are saved,
allowing users to revisit previous analyses.get_marker() Updates: Added extra
messages when searching for undefined cells in the database, improving
clarity for users.prepare_geo() Fix: Non-character ID
columns in GPL data are now converted to character to prevent
misinterpretation of numeric IDs. Expression data can now be read
directly from supplementary files without needing to download them
locally.plotORA() Fix: Fixed an issue where
the plot would fail to display a legend when mapping a variable to the
fill aesthetic.uniprot_id_map() on the latest macOS during example
usage.These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.