flywire_leaves()
cache
by @jefferis in https://github.com/natverse/fafbseg/pull/211flywire_cave_query()
set live=TRUE
if we provided a timestamp (e65e966)Full Changelog: https://github.com/natverse/fafbseg/compare/v0.15.1...v0.15.2
Full Changelog: https://github.com/natverse/fafbseg/compare/v0.15.0...v0.15.1
This contains substantial under the hood changes in flywire_cave_query()
which may be breaking in some instances, based on fixes of my
understanding of CAVE live queries.
Changes in flywire_partner_summary()
are also significant in allowing
considerably faster queries for multiple input neurons with cave
now the
default query method.
flywire_cave_query()
: add livelive, fix live query by @jefferis in https://github.com/natverse/fafbseg/pull/206flywire_partner_summary()
queries by @jefferis in https://github.com/natverse/fafbseg/pull/205Full Changelog: https://github.com/natverse/fafbseg/compare/v0.14.2...v0.15.0
This is a pre-release in preparation for a major (0.15.0) release and does feature some API changes in flywire_partner_summary()
(motivated by fanc/banc support in the coconatfly package).
flywire_partner_summary()
by @jefferis in https://github.com/natverse/fafbseg/pull/203flywire_cave_query()
query enhancements inc select_columns
to speed up partner queries by @jefferis in https://github.com/natverse/fafbseg/pull/202nucleus_table_info()
and friends by @jefferis in https://github.com/natverse/fafbseg/pull/204Full Changelog: https://github.com/natverse/fafbseg/compare/v0.14.1...v0.14.2
flywire_ids()
to accept any whitespace by @jefferis in https://github.com/natverse/fafbseg/pull/198flywire_ids()
to accept file argument by @jefferis in https://github.com/natverse/fafbseg/pull/199dr_fafbseg()
(1eabc55b)Full Changelog: https://github.com/natverse/fafbseg/compare/v0.14.0...v0.14.1
This major release provides support for the v783 materialisation that we expect to accompany the public release of the FlyWire dataset. It is also features numerous bug fixes/usability improvements.
Full Changelog: https://github.com/natverse/fafbseg/compare/v0.13.0...v0.14.0
Full Changelog: https://github.com/natverse/fafbseg/compare/v0.12.0...v0.13.0
flywire_connectome_data()
and flywire_partner_summary2()
and flywire_adjacency_matrix2()
) by @jefferis in https://github.com/natverse/fafbseg/pull/170flywire_ids()
(and most fafbseg functions that take ids as inputs) now supports full regular expressions to define id queries. For example /type:MBON0[24]
(fc9df01b69eb5c540bc0bb1ebec89aad4621c695).flywire_islatest()
has a cache option when using a specific version/timestamp (4b284d7d86b7e1a6da980c521a282fa693f71456). This means that flywire_updateids()
and flywire_partner_summary2()
can be very fast when using a fixed version.read_l2dp()
and read_l2skel()
flywire_raw2nm()
use vox dims if supplied by @jefferis in https://github.com/natverse/fafbseg/pull/169xform.ngscene()
by @jefferis in https://github.com/natverse/fafbseg/pull/168fafbseg.condaenv
option if you want to specify a non-standard miniconda environment.write_nginfo()
) by @jefferis in https://github.com/natverse/fafbseg/pull/173flytable_meta
as a simple interface to get flytable cell type information for a set of flywire_idsFull Changelog: https://github.com/natverse/fafbseg/compare/v0.11.1...v0.12.0
Full Changelog: https://github.com/natverse/fafbseg/compare/v0.11.0...v0.11.1
Full Changelog: https://github.com/natverse/fafbseg/compare/v0.10.0...v0.11.0
The key change is to react to a breaking change in the URL to access spine services.
Full Changelog: https://github.com/natverse/fafbseg/compare/v0.9.4...v0.10.0
flywire_cave_client()
flywire_cave_query()
(#122)flywire_updateids()
(#127)flywire_rootid()
flytable_*
functions including flytable_query()
flytable_list_rows()
flytable_login()
(#126)flywire_partners()
https://github.com/natverse/fafbseg/issues?q=closed%3A2021-09-20..2021-10-29+
This is a pre-release without full release notes, principally to support a new dependant package, fancr.
flywire_scene()
function for simple specification of flywire
neuroglancer scenes.ngl_segments<-()
to replace the segments in a scene
+.ngscene()
and -.ngscene()
to add or remove segments from a scene and as.character.ngscene()
to convert a scene object to a URL. See examples in ngl_segments()
and ngl_decode_scene()
for further details (#55, #58).flywire_set_token()
to support new cave and zetta tokens
(for FANC dataset) (#119)chunkedgraph_token()
function (#119)check_cloudvolume_reticulate
now supports a minimum versionmethod="cave"
option for flywire_partner_summary()
(#122).https://github.com/natverse/fafbseg/issues?q=closed%3A2020-12-27..2021-09-20+
The version change to 0.9 reflects some small but potentially breaking changes in user visible behaviour.
In particular ngl_segments()
now returns as character by default (to protect FlyWire ids, which cannot be expressed as 8 byte floating point values).
flywire_latestid()
function to map any (past) root id to the current root id. Specifically it it identifies the neuron inheriting the largest number of super voxels from the object specified by the input root id. ngl_segments()
return ids as character vector, rather than numeric, by default. Also exclude hidden ids by default (#50, #51).ngl_segments()
(and flywire_*) fns to accept URLs with surface
meshes. Previously these would cause an error (#52)ngl_encode_url()
so that all tested flywire scenes can be parsed in R and then converted back to valid URLs that can be opened in flywire.ai (#53)flywire_cloudvolume()
enables improved low-level support for flywire+cloudvolume (by exposing a reusable cloudvolume object) (#48)full list of closed issues on GitHub.
flywire_xyz2id()
supervoxel id mapping using spine service (#44)flywire_leaves()
function to find supervoxels in a neuron (#43)You can see the full list of closed issues on GitHub.
skeletor()
to skeletonise meshes especially flywire neurons
based on the Python package of the same name (#35).flywire_rootid()
(#37)flywire_xyz2id()
so that it can actually find supervoxels (as well as root ids) (#37)dr_fafbseg()
to give a status update on your installation (#36)read_cloudvolume_meshes()
can now accept flywire URLs (which will be expanded if necessary to define the segments to download)You can see the full list of closed issues on GitHub.
fafb2flywire()
during xform_brain
#30, #32, #33wire3d
#27fafb2flywire()
i.e. the inverse of the original FlyWire->FAFB transformation #26read_cloudvolume_meshes()
to work for flywire URLs #20ngl_segments()
needs to be able to expand flywire URLs #16You can see the full list of closed issues on GitHub.
read_cloudvolume_meshes()
for details including the required
Python setup.meshparty_skeletonize()
flywire2fafb()
to support for transforming neurons FlyWire->FAFB
(and with some loss of accuracy in the reverse) (#11)brainmaps_xyz2id()
reduce to 200 to reflect API
changes.choose_segmentation()
and with_segmentation()
to choose
default auto-segmentationbrainmaps_fetch()
a generic cache option and the ability to clear the
cache with brainmaps_clear_cache()
(#8)brainmaps_fetch()
a retry option to help with sporadic timeouts (#9)catmaid2ngl()
a chunksize option that can be used to reduce timeout
issues..onLoad
and retire fafbseg.divisor option
(now calculated automatically rather than being a user option)brainmaps_voxdims
function using cache mechanismdplyr::n()
in find_topn()
find_topn()
and read_top()
ability to return segments in
increasing (rather than decreasing) size order.ngl_encode_url()
when only one segment in scene (#5)catmaid2ngl()
function and methods, a high level approach to converting
URLs, neurons etc to representations based on Neuroglancer / brainmaps data.brainmaps_xyz()
has been updated to
brainmaps://772153499790:fafb_v14:fafb_v14_16nm_v00c_split3xfill2_flatreseg2_skeletons32
read.neurons.brainmaps()
brainmaps_skeleton()
and read.neurons.brainmaps()
now use a brainmaps
URI to specify the remote skeleton source.fafbseg.skeletonuri
to specify default remote skeleton
sourcebrainmaps_xyz2id()
a chunksize
argument to handle more query points
than the remote API will accept in a single call.read.neuron.brainmaps()
and read.neurons.brainmaps()
to read skeletons
into nat neuron objects over the web.brainmaps_skeleton()
function.find_merge_groups()
to return segment idscompare_ng_neuron()
: add pointsize and sample_dots argsread_brainmaps_meshes()
brainmaps_xyz2id()
(#2)brainmaps_auth()
, brainmaps_fetch()
to talk to brainmaps APIbrainmaps_xyz2id()
to convert arbitrary XYZ locations to the corresponding
segment id.find_merged_segments()
to return the merge groupsread_mergeinfo()
, make_merge_graph()
, merge_graph_components()
. These were
used to generate the fafbsegdata
package and could be useful in their own
right.find_merged_segments()
to find all the raw segments that are candidate
merges from the agglomeration runs. This functionality depends on the new
fafbsegdata package, which should be installed automatically as a suggested
dependency.fafb14_v00c_split3xfill2x_skeleton32nm512_nnconn75_thresh1000_sparse250
)read_segments()
datafrac
argument logic - was only reading largest fragmentdatafrac
only applies to files > minfilesize
skelsforsegments()
function with progress and use it in read_segments2()
zip_list
memoisation again but with a 5 min cache timeout - can speed
up read_segments2 considerably when there are multiple segments inside the
same zip file.read_segments2()
to read only the top n percent of skeletons by file sizengl_decode_scene()
function to parse neuroglancer URLs or JSON scene
specifications (in turn allowing these to be opened in CATMAID)open_fafb_ngl()
open_fafb_ngl
to produce coordinate string to paste into Neuroglancerngl_segments()
to extract segment ids from diverse scene specifications such
as URLs or JSON fragments copied from Neuroglancer web pageread_segments2()
(in memory zip extraction) recommendedfind_topn()
to find biggest segments in zip filesread_topn()
to cope with multiple zip filesfafb_v14_16nm_v00c_split3xfill2
)
that is compatible with the skeletons we have received.Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.