The goal of gardenr is provide simple functions of analyzing the data
from Gardener’s interneuron classification as well as to provide
examples of analyses of such data.
You can use the devtools package to install gardenr from Github.
devtools::install_github('ComputationalIntelligenceGroup/gardenr')The user needs to provide the path to the folder containing the data. These can be downloaded from https://figshare.com/s/8a761698160a675bb080.
folder <- '/home/bmihaljevic/code-gardener/gardener-experiment-data/data/'
library(tidyverse)
#> ── Attaching packages ───────────────────────────────────────────────────────────────────────────────────────────────────────────────── tidyverse 1.2.1 ──
#> ✔ ggplot2 3.2.0 ✔ purrr 0.3.2
#> ✔ tibble 2.1.3 ✔ dplyr 0.8.1
#> ✔ tidyr 0.8.3 ✔ stringr 1.4.0
#> ✔ readr 1.3.1 ✔ forcats 0.4.0
#> ── Conflicts ──────────────────────────────────────────────────────────────────────────────────────────────────────────────────── tidyverse_conflicts() ──
#> ✖ dplyr::filter() masks stats::filter()
#> ✖ dplyr::lag() masks stats::lag()gardenr functions return the data for all 320 cells. There are
functions for the classification labels get_all_labels, metadata
get_all_meta, and alternative type names and definitions
get_alternative_types.
library(gardenr)
#> Warning: replacing previous import 'magrittr::extract' by 'tidyr::extract'
#> when loading 'gardenr'
#>
#> Attaching package: 'gardenr'
#> The following object is masked from 'package:readr':
#>
#> read_csv
annotations <- get_all_labels(folder )
summary(annotations)
#> annotator neuron F1 F2
#> 1 : 320 1 : 48 : 151 : 175
#> 2 : 320 2 : 48 intralaminar:3895 intracolumnar:7660
#> 3 : 320 4 : 48 None :1326 None :1326
#> 4 : 320 5 : 48 translaminar:8970 transcolumnar:5181
#> 5 : 320 6 : 48
#> 6 : 320 7 : 48
#> (Other):12422 (Other):14054
#> F3 F4 F5
#> : 254 : 375 common type :2934
#> centered :5649 ascending :2390 common basket:2839
#> displaced:7113 both :1352 large basket :1941
#> None :1326 descending:2543 Martinotti :1563
#> None :7682 None :1326
#> neurogliaform:1032
#> (Other) :2707
#> F6 other complete
#> characterized :13016 None :13346 Mode :logical
#> uncharacterized: 1326 columnar basket: 70 FALSE:482
#> unknown : 68 TRUE :13860
#> bushy cell : 59
#> bitufted : 40
#> double bouquet : 40
#> (Other) : 719alternative <- get_alternative_types(folder)
summary(alternative)
#> annotator type definition
#> 23 :74 bipolar : 5 :163
#> 27 :68 bitufted : 5 see above. : 12
#> 14 :38 double bouquet : 4 as it says. : 6
#> 7 :11 bipolar? : 3 see above : 4
#> 30 :10 inverted arcade: 3 as above. : 2
#> 18 : 9 bitufted? : 2 deleted from the final list.: 2
#> (Other):59 (Other) :247 (Other) : 80metadata <- get_all_meta(folder)
summary(metadata)
#> neuron neuromorpho.name species area
#> 1 : 1 None : 79 Cat : 10 Area not reported: 1
#> 2 : 1 020227-slice1: 1 Human : 11 Auditory : 1
#> 3 : 1 020315-2-ST : 1 Monkey: 68 Frontal : 8
#> 4 : 1 020315-3-ST : 1 Mouse : 78 Prefrontal : 33
#> 5 : 1 020515-2-NPY : 1 Rabbit: 2 Somatosensory :184
#> 6 : 1 020530-2-NPY : 1 Rat :151 Temporal : 9
#> (Other):314 (Other) :236 Visual : 84
#> layer rotated archive
#> II : 4 Mode :logical Gonzalez-Burgos: 15
#> III : 16 FALSE:310 Helmstaedter : 43
#> II/III :136 TRUE :10 Markram :104
#> IV : 98 Yuste : 78
#> not reported: 30 NA's : 80
#> V : 33
#> VI : 3
#> original.type
#> Not reported :79
#> Basket cell :57
#> Martinotti cell :33
#> Somatostatin (SOM) containing cell:18
#> Bitufted cell :12
#> (Other) :41
#> NA's :80
#> paper
#> Anatomical, physiological, molecular and circuit properties of nest basket cells in the developing somatosensory cortex :104
#> The relation between dendritic geometry, electrical excitability, and axonal projections of L2/3 Interneurons in rat barrel cortex.: 43
#> Internal dynamics determine the cortical response to thalamic stimulation : 17
#> Correlation between axonal morphologies and synaptic input kinetics of interneurons from mouse visual cortex : 16
#> Calcium microdomains in aspiny dendrites : 15
#> (Other) :104
#> NA's : 21To get data on a specific cell or annotator, we only need filter by id:
metadata %>% filter(neuron == 12)
#> neuron neuromorpho.name species area layer rotated
#> 1 12 03-27-01-3wideArbor Monkey Prefrontal II/III FALSE
#> archive original.type
#> 1 Gonzalez-Burgos Basket cell
#> paper
#> 1 Cluster analysis-based physiological classification and morphological properties of inhibitory neurons in layers 2-3 of monkey dorsolateral prefrontal cortex
head(annotations %>% filter(neuron == 12))
#> annotator neuron F1 F2 F3 F4
#> 1 1 12 translaminar transcolumnar centered None
#> 2 2 12 translaminar transcolumnar centered None
#> 3 3 12 translaminar transcolumnar displaced descending
#> 4 4 12 translaminar transcolumnar displaced both
#> 5 5 12 translaminar transcolumnar centered None
#> 6 6 12 translaminar transcolumnar centered None
#> F5 F6 other complete
#> 1 large basket characterized None TRUE
#> 2 large basket characterized None TRUE
#> 3 large basket characterized None TRUE
#> 4 large basket characterized None TRUE
#> 5 large basket characterized None TRUE
#> 6 large basket characterized None TRUEhead(annotations %>% filter(annotator == 2))
#> annotator neuron F1 F2 F3 F4
#> 1 2 1 intralaminar intracolumnar centered None
#> 2 2 2 translaminar transcolumnar displaced both
#> 3 2 3 translaminar intracolumnar displaced ascending
#> 4 2 4 intralaminar intracolumnar centered None
#> 5 2 5 translaminar transcolumnar centered None
#> 6 2 6 intralaminar intracolumnar centered None
#> F5 F6 other complete
#> 1 neurogliaform characterized None TRUE
#> 2 chandelier characterized None TRUE
#> 3 Martinotti characterized None TRUE
#> 4 common type characterized None TRUE
#> 5 common basket characterized None TRUE
#> 6 neurogliaform characterized None TRUE
alternative %>% filter(annotator == 2)
#> [1] annotator type definition
#> <0 rows> (or 0-length row.names)It is easy to combine the three data frames using joins, and then
compute summaries with tidyverse functions. For example, the
alternative type definition to an annotation
head(left_join(annotations, alternative, by = c("other" = "type", "annotator" = "annotator")))
#> Warning: Column `other`/`type` joining factors with different levels,
#> coercing to character vector
#> Warning: Column `annotator` joining factors with different levels, coercing
#> to character vector
#> annotator neuron F1 F2 F3 F4
#> 1 1 1 intralaminar intracolumnar centered None
#> 2 1 2 translaminar transcolumnar displaced descending
#> 3 1 3 translaminar transcolumnar displaced ascending
#> 4 1 4 translaminar intracolumnar centered None
#> 5 1 5 translaminar transcolumnar centered None
#> 6 1 6 intralaminar intracolumnar centered None
#> F5 F6 other complete definition
#> 1 neurogliaform characterized None TRUE <NA>
#> 2 chandelier characterized None TRUE <NA>
#> 3 Martinotti characterized None TRUE <NA>
#> 4 common basket characterized None TRUE <NA>
#> 5 large basket characterized None TRUE <NA>
#> 6 neurogliaform characterized None TRUE <NA>The primary key in alternative consits of annotator and type and
hence we need both in the join.
There is a utility function to return the counts for all categories, with a single entry per neuron.
counts <- get_all_counts(folder)
#> Warning: attributes are not identical across measure variables;
#> they will be dropped
head(counts)
#> # A tibble: 6 x 22
#> neuron intralaminar translaminar intracolumnar transcolumnar centered
#> <fct> <int> <int> <int> <int> <int>
#> 1 1 44 2 44 0 44
#> 2 2 21 26 29 17 4
#> 3 3 1 45 36 8 1
#> 4 4 9 28 34 2 33
#> 5 5 4 43 8 39 30
#> 6 6 45 1 45 0 44
#> # … with 16 more variables: displaced <int>, ascending <int>, both <int>,
#> # descending <int>, arcade <int>, `Cajal-Retzius` <int>,
#> # chandelier <int>, `common basket` <int>, `common type` <int>,
#> # `horse-tail` <int>, `large basket` <int>, Martinotti <int>,
#> # neurogliaform <int>, other <int>, characterized <int>,
#> # uncharacterized <int>We can join the above with metadata to get thorough information for each cell.
meta_count <- left_join(counts, metadata, by = 'neuron') %>% select(-paper, -original.type, -archive, -neuromorpho.name)
head(meta_count)
#> # A tibble: 6 x 26
#> neuron intralaminar translaminar intracolumnar transcolumnar centered
#> <fct> <int> <int> <int> <int> <int>
#> 1 1 44 2 44 0 44
#> 2 2 21 26 29 17 4
#> 3 3 1 45 36 8 1
#> 4 4 9 28 34 2 33
#> 5 5 4 43 8 39 30
#> 6 6 45 1 45 0 44
#> # … with 20 more variables: displaced <int>, ascending <int>, both <int>,
#> # descending <int>, arcade <int>, `Cajal-Retzius` <int>,
#> # chandelier <int>, `common basket` <int>, `common type` <int>,
#> # `horse-tail` <int>, `large basket` <int>, Martinotti <int>,
#> # neurogliaform <int>, other <int>, characterized <int>,
#> # uncharacterized <int>, species <fct>, area <fct>, layer <fct>,
#> # rotated <lgl>There is a utility function doing just that:
head(get_all_counts_meta(folder) )
#> Warning: attributes are not identical across measure variables;
#> they will be dropped
#> # A tibble: 6 x 26
#> neuron intralaminar translaminar intracolumnar transcolumnar centered
#> <fct> <int> <int> <int> <int> <int>
#> 1 1 44 2 44 0 44
#> 2 2 21 26 29 17 4
#> 3 3 1 45 36 8 1
#> 4 4 9 28 34 2 33
#> 5 5 4 43 8 39 30
#> 6 6 45 1 45 0 44
#> # … with 20 more variables: displaced <int>, ascending <int>, both <int>,
#> # descending <int>, arcade <int>, `Cajal-Retzius` <int>,
#> # chandelier <int>, `common basket` <int>, `common type` <int>,
#> # `horse-tail` <int>, `large basket` <int>, Martinotti <int>,
#> # neurogliaform <int>, other <int>, characterized <int>,
#> # uncharacterized <int>, species <fct>, area <fct>, layer <fct>,
#> # rotated <lgl>These data are also available in a different format, which is a bit easier to analyze
annot_tidy <- make_annotations_tidy(annotations)
#> Warning: attributes are not identical across measure variables;
#> they will be dropped
head(annot_tidy )
#> annotator neuron feature value
#> 1 1 1 F1 intralaminar
#> 2 1 2 F1 translaminar
#> 3 1 3 F1 translaminar
#> 4 1 4 F1 translaminar
#> 5 1 5 F1 translaminar
#> 6 1 6 F1 intralaminar- How many partially labelled cells?
table(annotations$complete)
#>
#> FALSE TRUE
#> 482 13860- Get raw frequencies of the different categories
annot_tidy <- make_annotations_tidy(annotations)
#> Warning: attributes are not identical across measure variables;
#> they will be dropped
annot_tidy %>% group_by(feature, value) %>% tally()
#> # A tibble: 31 x 3
#> # Groups: feature [6]
#> feature value n
#> <chr> <fct> <int>
#> 1 F1 "" 151
#> 2 F1 None 1326
#> 3 F1 intralaminar 3895
#> 4 F1 translaminar 8970
#> 5 F2 "" 175
#> 6 F2 None 1326
#> 7 F2 intracolumnar 7660
#> 8 F2 transcolumnar 5181
#> 9 F3 "" 254
#> 10 F3 None 1326
#> # … with 21 more rows
# library(ggplot2)
ggplot(annot_tidy, aes(x = value, color = feature)) + geom_bar() + theme(axis.text.x = element_text(angle = 90, hjust = 1))
- Only consider cells with above > 30 neurocientists agreeing on a specific label
annot_tidy <- make_annotations_tidy(annotations)
#> Warning: attributes are not identical across measure variables;
#> they will be dropped
above30 <- annot_tidy %>% group_by(neuron, feature, value) %>% tally() %>% filter(n > 30)
above30 %>% group_by(feature) %>% summarize(n_distinct(neuron))
#> # A tibble: 6 x 2
#> feature `n_distinct(neuron)`
#> <chr> <int>
#> 1 F1 237
#> 2 F2 240
#> 3 F3 216
#> 4 F4 219
#> 5 F5 57
#> 6 F6 305- Restrict analysis to unrotated, monkey cells
monkey <- metadata %>% filter(species == 'Monkey')
counts <- get_all_counts_meta(folder)
#> Warning: attributes are not identical across measure variables;
#> they will be dropped
counts %>% filter(species == 'Monkey')
#> # A tibble: 68 x 26
#> neuron intralaminar translaminar intracolumnar transcolumnar centered
#> <fct> <int> <int> <int> <int> <int>
#> 1 1 44 2 44 0 44
#> 2 6 45 1 45 0 44
#> 3 9 4 43 8 38 30
#> 4 12 9 35 1 44 35
#> 5 14 1 43 45 0 27
#> 6 22 2 39 41 1 4
#> 7 27 18 27 44 1 39
#> 8 28 1 40 39 3 1
#> 9 33 34 9 41 1 31
#> 10 37 37 0 36 0 30
#> # … with 58 more rows, and 20 more variables: displaced <int>,
#> # ascending <int>, both <int>, descending <int>, arcade <int>,
#> # `Cajal-Retzius` <int>, chandelier <int>, `common basket` <int>,
#> # `common type` <int>, `horse-tail` <int>, `large basket` <int>,
#> # Martinotti <int>, neurogliaform <int>, other <int>,
#> # characterized <int>, uncharacterized <int>, species <fct>, area <fct>,
#> # layer <fct>, rotated <lgl>- 27 out of 48 neuroscientists provided alternative type names for the cells
length(unique(alternative$annotator))
#> [1] 27
unique(alternative$annotator)
#> [1] 1 4 7 9 13 14 15 18 19 23 24 25 26 27 29 30 31 34 35 36 37 39 40
#> [24] 41 43 44 48
#> 27 Levels: 1 4 7 9 13 14 15 18 19 23 24 25 26 27 29 30 31 34 35 36 ... 48- How many alternative types per neuroscientist? How many other cells per neuroscientist?
types <- alternative %>% group_by(annotator) %>% tally()
other <- annotations %>% group_by(annotator) %>% filter(F5 == 'other') %>% tally()
ggplot(data.frame(types = types$n, cells = other$n), aes(x = types, y = cells)) + geom_point()
- Words used in alternative types and definitions:
- The most common word is `bitufted’
library(tidytext)
library(wordcloud)
#> Loading required package: RColorBrewer
alt <- alternative
alt$type <- as.character(alternative$type )
alt <- alt %>% unnest_tokens(type, type)
alt %>% count(type, sort = TRUE)
#> # A tibble: 271 x 2
#> type n
#> <chr> <int>
#> 1 bitufted 44
#> 2 arbor 38
#> 3 bipolar 38
#> 4 with 36
#> 5 be 35
#> 6 could 30
#> 7 martinotti 29
#> 8 cell 28
#> 9 a 26
#> 10 axonal 25
#> # … with 261 more rows
alt %>% count(type, sort = TRUE) %>% with(wordcloud(type, n, random.order = FALSE, max.words = 50 , colors= brewer.pal(8,"Dark2")))
alt <- alternative
alt$definition <- as.character(alternative$definition )
alt <- alt %>% unnest_tokens(definition, definition)
alt <- alt %>% anti_join(get_stopwords(), by = c("definition" = "word"))
alt %>% count(definition, sort = TRUE)
#> # A tibble: 427 x 2
#> definition n
#> <chr> <int>
#> 1 axonal 37
#> 2 cells 30
#> 3 cell 25
#> 4 see 22
#> 5 layer 20
#> 6 soma 18
#> 7 arbor 15
#> 8 domain 15
#> 9 layers 14
#> 10 dendritic 13
#> # … with 417 more rows
alt %>% count(definition, sort = TRUE) %>% with(wordcloud(definition, n, random.order = FALSE, max.words = 50 , colors= brewer.pal(8,"Dark2")))