profile
viewpoint
Gregory Jefferis jefferis MRC Laboratory of Molecular Biology Cambridge, United Kingdom http://www2.mrc-lmb.cam.ac.uk/group-leaders/h-to-m/gregory-jefferis/

jefferis/AnalysisSuite 6

R Neuron Analysis Suite provides functions to read/write/analyse neuroanatomical data

jefferis/cmtk 5

Unofficial git clone of Computational Morphometry Toolkit (CMTK)

jefferis/AnalysisSuiteBin 2

These are additional perl/shell scripts that are intended to support my AnalysisSuite package

alexanderbates/deformetricar 1

R wrapper for deformetrica registration tool

jefferis/alphashape3d 1

The package alphashape3d presents the implementation in R of the alpha-shape of a finite set of points in the three-dimensional space.

jefferis/bmp 1

A simple R package to read Windows BMP format images

bocklab/skeleton_server 0

Hack to view CATMAID skeletons in Neuroglancer

flyconnectome/temca2data 0

Public code for Zheng, Lauritzen et al. (2017)

push eventnatverse/fafbseg

alexanderbates

commit sha 1e01ec8535afdc0729ddae284108602add3f0c52

Update to .Rmd file, reminding to use latest cloudvolume

view details

push time in 11 minutes

PR opened natverse/fafbseg

Cloudvol option

A very small set of changes to:

  • Carry cloudvolume.url argument to flywire_partner_summary, as we are using it elsewhere
  • Put dplyr:: in front of some dplyr functions, as I was getting thrown while testing by a namespace clash
+13 -11

0 comment

2 changed files

pr created time in 44 minutes

issue commentnatverse/fafbseg

more informative error for flywire_xyz2id when coordinate space might be wrong

@tatsuookubo this was the issue we had ealier

alexanderbates

comment created time in an hour

push eventnatverse/fafbseg

alexanderbates

commit sha bb9ff5bf31b848cf9268edef6236a16cb5dbd243

Properly carry through cloudvolume.url argument internally * oops

view details

alexanderbates

commit sha 64ca397e625b91d114e9dcf34bd1e6aca1bf4fe7

Use cloudvolume.url=cloudvolume.url with internal flywire_partner_summary

view details

alexanderbates

commit sha f984c808ce362d3a2473b2ae70038fc69a19c2f3

roxygenise()

view details

push time in an hour

create barnchnatverse/fafbseg

branch : cloudvol-option

created branch time in an hour

push eventnatverse/fafbseg

alexanderbates

commit sha 08f0e5216de1a1e650293e2f96a98c9de018a83e

return unmodified neuron if no synapses could be retreived

view details

push time in an hour

issue openednatverse/fafbseg

more informative error for

This works:

# get ID
flywire.xyz = "108847.975,62454.2775,3811"
flywire.id = fafbseg::flywire_xyz2id(nat::xyzmatrix(flywire.xyz), rawcoords = TRUE)

However, earlier today I forgot that the default was not rawcoords = TRUE. If I do it with FALSE, I get this error:

flywire.xyz = "108847.975,62454.2775,3811"
flywire.id = fafbseg::flywire_xyz2id(nat::xyzmatrix(flywire.xyz), rawcoords = FALSE)
> Error in parse_url(url) : length(url) == 1 is not TRUE

And didn't immediately realise what was wrong. I also happened to get two separate inquiries over the last few days about why this function was not working, both times it was the same issue.

More informative error message needed?

created time in an hour

delete branch natverse/fafbseg

delete branch : feature/flywire-partners-both

delete time in an hour

push eventnatverse/fafbseg

Gregory Jefferis

commit sha b0c002c0661d6ac69d25d7d8f14f2a74c205a6f8

flywire_partners now can take partners = "both"

view details

Gregory Jefferis

commit sha d733e63c4b94500219be37067146d9aba5ae0cfc

remove duplicate code in test

view details

Gregory Jefferis

commit sha 773de5887339b99b45ba2ff620b5679822edc822

add test for flywire_partners both

view details

Gregory Jefferis

commit sha 65694d33f1db05b0a53ae2c6de4bee8214c3c28b

speed up flywire_partners both for big neurons * don't try to embed ids in query

view details

Gregory Jefferis

commit sha 43c81e98c0cbad7a75671e346a646b01384504a3

some small optimisations to the sqlite processing * delay sql evaluation in case we are going to add details * better handling of sorting * ... while avoiding strange choice of sqlite optimiser

view details

Gregory Jefferis

commit sha 86daf3a6533402eb2bb31d6c8f2920d52cf5d94b

additional tests for sqlite synapse dbs

view details

Gregory Jefferis

commit sha e7b4fe91ea854d6f25c1870162078996a333b655

roxygenise

view details

Alexander Bates

commit sha eba061a162be68cecba3d9152e86ee001fd80c53

Merge pull request #77 from natverse/feature/flywire-partners-both Feature/flywire partners both

view details

push time in an hour

PR merged natverse/fafbseg

Reviewers
Feature/flywire partners both

based on initial code from @alexanderbates in #72 after some cleanup and optimisation of the sqlite query (which could be very slow (mins) for large neurons)

+76 -35

2 comments

3 changed files

jefferis

pr closed time in an hour

pull request commentnatverse/fafbseg

Feature/flywire partners both

With Greg's change, much faster, for when partners = "both" which we need when getting all input and output synapses for a neuron skeleton. This is then useful for splitting into axon/dendrite.

jefferis

comment created time in an hour

Pull request review commentnatverse/fafbseg

Add flywire synapses to flywire neurons

 flywire_partners <- function(rootid, partners=c("outputs", "inputs"),     colnames(resdf) <- c("offset", 'pre_svid', "post_svid", "scores", "cleft_scores")     # we can get the same row appearing twice for autapses     resdf <- filter(resdf, !duplicated(.data$offset))-    resdf <- if(partners=='outputs')-      filter(resdf, .data$pre_svid%in%svids) else filter(resdf, .data$post_svid%in%svids)-+    if (partners == "outputs"){+      resdf <-  filter(resdf, .data$pre_svid %in% svids)+    }else if (partners == "inputs"){+      resdf <-  filter(resdf, .data$post_svid %in% svids)+    }   } else {-    df <- if(partners=="outputs")-      tibble::tibble(pre_svid=svids) else tibble::tibble(post_svid=svids)-    res=dplyr::inner_join(flywireids, df,-                          by=ifelse(partners=="outputs", "pre_svid", "post_svid"),-                          copy=TRUE, auto_index=TRUE, )+    if(partners == "both"){+      res = dplyr::filter(flywireids,+                          .data$pre_svid%in%svids|.data$post_svid%in%svids)

Greg says:

https://github.com/natverse/fafbseg/pull/73 merged https://github.com/natverse/fafbseg/pull/77 pending your review but I think good to merge now I will put up the last PR now. Note that there was a significant issue in 77 which was resulting in very slow sqlite queries – the final as.data.frame step triggering the sqlite query to run quickly got extremely slow for large neurons.

alexanderbates

comment created time in an hour

pull request commentnatverse/nat

Basic implementation of monolithic DB1 format for neuronlistfh

I think when the lock file is left, something has errored, and the datafile is not updated as I expected. I have not tested much though, because I only encountered this error a few times and both times did not investigate it thoroughly. I have not had it again since commenting here.

jefferis

comment created time in an hour

issue commentgoogle/neuroglancer

define get_skeleton() in SkeletonSource for custom data

Turns out the skeleton example was broken --- I just fixed it.

get_skeleton is called by the Python Neuroglancer code when the client requests a skeleton --- your own code doesn't need to call it directly.

hash-ir

comment created time in 2 hours

startedAppsilon/shiny.router

started time in 2 hours

startedAppsilon/shiny.semantic

started time in 2 hours

push eventgoogle/neuroglancer

Jeremy Maitin-Shepard

commit sha d91a20a00b6809eaa7487a04eebaf4f2c9f2d94a

fix(python): add SkeletonSource error checking and fix skeleton example Fixes #277

view details

push time in 2 hours

issue closedgoogle/neuroglancer

define get_skeleton() in SkeletonSource for custom data

I am running a skeleton example on custom data inspired by example_skeleton.py. I implemented the get_skeleton() method of SkeletonSource but I didn't find its use anywhere in the example. I want to know how I can forward my custom skeleton data to the viewer source?

In the skeleton example, the dummy segmentation data is provided in a LocalVolume and a SkeletonSource object is just instantiated without using the get_skeleton() method.

with viewer.txn() as s:
    s.layers.append(
        name='a',
        layer=neuroglancer.SegmentationLayer(
            source=[
                neuroglancer.LocalVolume(
                    data=segmentation, # here
                    dimensions=dimensions,
                ),
                SkeletonSource(dimensions), # here
            ],
            skeleton_shader='void main() { emitRGB(colormapJet(affinity)); }',
            selected_alpha=0,
            not_selected_alpha=0,
            segments=[395750],
        ))

closed time in 2 hours

hash-ir

issue commentVirtualFlyBrain/VFB_connect

EPIC: Build API wrapper for VFB_connect

Version1 released https://github.com/VirtualFlyBrain/VFB_connect_api Missing items are:

  • get_term_info: How to use type and filter parameters?
  • get_subclasses: a sample usage (query string example). How to use filter
  • get_instances: How to use filters?
  • get_term_info_by_xref: How to use filters
dosumis

comment created time in 2 hours

push eventVirtualFlyBrain/VFB_connect_api

Huseyin Kir

commit sha a60a53880d236fa42987acffd7dad05ab70970f7

Update README.md

view details

push time in 3 hours

push eventVirtualFlyBrain/VFB_connect_api

Huseyin Kir

commit sha 96d903ce4782d55c0a7648500c78602a7460411a

Update README.md

view details

push time in 3 hours

push eventVirtualFlyBrain/VFB_connect_api

Rob Court

commit sha af45a83cbd7247f67136775365ad9192d56803c0

seeing what debug says

view details

Rob Court

commit sha 6bb52cac18024e77c1f755bfb990b286dd6b9fea

opening permissions

view details

Rob Court

commit sha 3ca0b38abfa334472df278fe3c6112fe29a02307

Revert "seeing what debug says" This reverts commit af45a83cbd7247f67136775365ad9192d56803c0.

view details

Huseyin Kir

commit sha 29d429f12d24a74492b8025fd3197cad555c5d95

Merge pull request #1 from VirtualFlyBrain/debug opening permissions

view details

push time in 3 hours

push eventVirtualFlyBrain/VFB_connect_api

Rob Court

commit sha 3ca0b38abfa334472df278fe3c6112fe29a02307

Revert "seeing what debug says" This reverts commit af45a83cbd7247f67136775365ad9192d56803c0.

view details

push time in 3 hours

push eventVirtualFlyBrain/VFB_connect_api

Rob Court

commit sha 6bb52cac18024e77c1f755bfb990b286dd6b9fea

opening permissions

view details

push time in 3 hours

push eventVirtualFlyBrain/VFB_connect_api

Rob Court

commit sha af45a83cbd7247f67136775365ad9192d56803c0

seeing what debug says

view details

push time in 3 hours

create barnchVirtualFlyBrain/VFB_connect_api

branch : debug

created branch time in 3 hours

push eventVirtualFlyBrain/VFB_connect_api

Huseyin Kir

commit sha 00002966c597a46139c2534e799f87039ff59c0c

Update README.md

view details

push time in 4 hours

more