1
00:00:04,960 --> 00:00:19,999
[Music]

2
00:00:20,519 --> 00:00:26,640
hello and welcome to our next talk here

3
00:00:23,800 --> 00:00:29,320
at Pon we're going to hear from Jack

4
00:00:26,640 --> 00:00:31,079
reichelt about walking the Tree of Life

5
00:00:29,320 --> 00:00:34,439
Adventures in

6
00:00:31,079 --> 00:00:34,439
phenetics thank you

7
00:00:36,600 --> 00:00:44,280
Jack good run and thanks for coming to

8
00:00:39,440 --> 00:00:48,600
my talk so I am Jack reichelt wearer of

9
00:00:44,280 --> 00:00:51,120
too many hats um I'm the CTO and

10
00:00:48,600 --> 00:00:53,359
co-founder of a startup called Kumo

11
00:00:51,120 --> 00:00:55,600
study uh we make a study management tool

12
00:00:53,359 --> 00:00:56,520
for students with a focus on students

13
00:00:55,600 --> 00:00:59,519
with

14
00:00:56,520 --> 00:01:00,680
ADHD uh we're doing U we have sort of a

15
00:00:59,519 --> 00:01:05,280
prototype version we're launching the

16
00:01:00,680 --> 00:01:08,159
real version 2 come marchish next year

17
00:01:05,280 --> 00:01:10,240
um I am big into Computer Science

18
00:01:08,159 --> 00:01:12,439
Education this is actually my first Pon

19
00:01:10,240 --> 00:01:16,119
talk that is not part of the education

20
00:01:12,439 --> 00:01:19,520
track and most relevant to this talk I

21
00:01:16,119 --> 00:01:22,360
am the holder of a wide variety of ADHD

22
00:01:19,520 --> 00:01:24,360
given interests so now that you know a

23
00:01:22,360 --> 00:01:26,960
little bit about me we're going to go

24
00:01:24,360 --> 00:01:28,479
back to that title slide uh because this

25
00:01:26,960 --> 00:01:30,680
talk has a subtitle which was mentioned

26
00:01:28,479 --> 00:01:32,680
the adventures in phog gentics it also

27
00:01:30,680 --> 00:01:35,880
has a sub subtitle which was more

28
00:01:32,680 --> 00:01:37,439
recently added and not on the program uh

29
00:01:35,880 --> 00:01:40,640
how it all went

30
00:01:37,439 --> 00:01:42,799
wrong so this project didn't end up

31
00:01:40,640 --> 00:01:44,880
where I expected it to but along the way

32
00:01:42,799 --> 00:01:46,880
I found some interesting stuff I learned

33
00:01:44,880 --> 00:01:49,479
a thing or two and I hope there's

34
00:01:46,880 --> 00:01:51,240
something you'll find of value here so I

35
00:01:49,479 --> 00:01:52,680
promise that we will get to the computer

36
00:01:51,240 --> 00:01:54,719
science but first we're going to take a

37
00:01:52,680 --> 00:01:56,200
detour through some other Sciences in

38
00:01:54,719 --> 00:01:59,520
order to understand what I was trying to

39
00:01:56,200 --> 00:02:04,439
do and why so let's start by clarifying

40
00:01:59,520 --> 00:02:06,759
what is phenetics so phog gentics is the

41
00:02:04,439 --> 00:02:08,479
study of life and evolution using

42
00:02:06,759 --> 00:02:11,200
genetics specifically it's not just

43
00:02:08,479 --> 00:02:14,080
saying these two organ organisms seem

44
00:02:11,200 --> 00:02:16,760
alike so we'll group them it's using

45
00:02:14,080 --> 00:02:20,360
molecular analysis of DNA and

46
00:02:16,760 --> 00:02:22,640
empirically cataloging and categorizing

47
00:02:20,360 --> 00:02:23,879
organisms remember back in school in

48
00:02:22,640 --> 00:02:26,599
your science class when you learned

49
00:02:23,879 --> 00:02:29,400
about taxonomy the whole domain Kingdom

50
00:02:26,599 --> 00:02:30,480
philm yada yada yada that's what I'm

51
00:02:29,400 --> 00:02:33,200
going to call

52
00:02:30,480 --> 00:02:35,959
Vibes based

53
00:02:33,200 --> 00:02:38,319
classification as an example giant

54
00:02:35,959 --> 00:02:41,560
pandas were named as such because of

55
00:02:38,319 --> 00:02:43,760
their obvious similarities to Red Pandas

56
00:02:41,560 --> 00:02:47,879
which were just called pandas at the

57
00:02:43,760 --> 00:02:50,760
time um in actuality giant pandas are a

58
00:02:47,879 --> 00:02:53,239
type of bear uh and red pandas are more

59
00:02:50,760 --> 00:02:55,959
closely related to skunks and

60
00:02:53,239 --> 00:02:57,720
raccoons now we later corrected this and

61
00:02:55,959 --> 00:03:00,280
realized that giant pandas are just

62
00:02:57,720 --> 00:03:03,560
really weird bears but the whole process

63
00:03:00,280 --> 00:03:07,280
was still pretty

64
00:03:03,560 --> 00:03:08,920
nonscientific also it's a bit weird to

65
00:03:07,280 --> 00:03:11,680
assume that every organism just has

66
00:03:08,920 --> 00:03:15,599
eight layers right like if each layer is

67
00:03:11,680 --> 00:03:15,599
a Divergence from other

68
00:03:15,760 --> 00:03:20,120
organisms that doesn't make a lot of

69
00:03:17,959 --> 00:03:23,040
sense has every organism only

70
00:03:20,120 --> 00:03:25,879
experienced seven divergences and what

71
00:03:23,040 --> 00:03:28,599
about the earliest organisms how had

72
00:03:25,879 --> 00:03:30,920
they experienced seven divergences or

73
00:03:28,599 --> 00:03:34,879
did they not have a bunch of these was

74
00:03:30,920 --> 00:03:38,439
it just species blank for all the

75
00:03:34,879 --> 00:03:40,400
rest and if it's that one then how long

76
00:03:38,439 --> 00:03:42,720
until we have to add a new tier and

77
00:03:40,400 --> 00:03:44,519
where does it fit in is are you going to

78
00:03:42,720 --> 00:03:45,959
add something below species I mean I

79
00:03:44,519 --> 00:03:48,840
guess that's kind of subspecies but

80
00:03:45,959 --> 00:03:51,400
that's not even on this list so taxonomy

81
00:03:48,840 --> 00:03:54,120
is a bit weird maybe it's not very

82
00:03:51,400 --> 00:03:58,319
good so we're going to use

83
00:03:54,120 --> 00:04:00,319
claes what is a clade a clade is

84
00:03:58,319 --> 00:04:04,640
basically just a group

85
00:04:00,319 --> 00:04:06,599
nothing all to it each CLA can have uh

86
00:04:04,640 --> 00:04:09,640
other species within it or if you're

87
00:04:06,599 --> 00:04:11,840
generalizing outside of uh organisms

88
00:04:09,640 --> 00:04:15,360
then just Leaf nodes or it can have

89
00:04:11,840 --> 00:04:17,079
other clades within it and it's clades

90
00:04:15,360 --> 00:04:19,639
all the way up just every successive

91
00:04:17,079 --> 00:04:21,880
tear up is always called a CLA humans of

92
00:04:19,639 --> 00:04:25,960
the species sapiens which is inside the

93
00:04:21,880 --> 00:04:29,160
CLA homo which is inside the CLA hmin I

94
00:04:25,960 --> 00:04:32,160
inside h a and so on all the way up to

95
00:04:29,160 --> 00:04:35,199
the Cade biota which contains all

96
00:04:32,160 --> 00:04:38,000
organisms for the last 4.2 billion years

97
00:04:35,199 --> 00:04:41,680
or so um there were actually some really

98
00:04:38,000 --> 00:04:43,400
early ones that weren't in biota um they

99
00:04:41,680 --> 00:04:47,759
all died out there's none of them alive

100
00:04:43,400 --> 00:04:49,639
everything came from one branch now so

101
00:04:47,759 --> 00:04:51,440
these clades in contrast to the taxonomy

102
00:04:49,639 --> 00:04:53,240
groupings are

103
00:04:51,440 --> 00:04:55,880
science-based they're formed using

104
00:04:53,240 --> 00:04:58,600
genetic analysis and the knowledge of

105
00:04:55,880 --> 00:05:00,160
the mutation rate of DNA so you can

106
00:04:58,600 --> 00:05:03,199
analyze the genomes of different

107
00:05:00,160 --> 00:05:05,639
organisms and tally up the differences

108
00:05:03,199 --> 00:05:07,320
and then count how many there are and

109
00:05:05,639 --> 00:05:09,320
that tells you how long ago they must

110
00:05:07,320 --> 00:05:12,039
have diverged you can use these

111
00:05:09,320 --> 00:05:15,240
divergences in order to plot them all

112
00:05:12,039 --> 00:05:15,240
out and categorize them

113
00:05:15,360 --> 00:05:20,400
all so what is the point of all of this

114
00:05:18,360 --> 00:05:24,440
why ultimately would we bother doing

115
00:05:20,400 --> 00:05:26,720
this well for starters it's more correct

116
00:05:24,440 --> 00:05:28,600
uh greatness is good and important in

117
00:05:26,720 --> 00:05:31,120
science it is worth it just for the

118
00:05:28,600 --> 00:05:32,880
better understanding that comes with it

119
00:05:31,120 --> 00:05:35,039
but it can also help us in more concrete

120
00:05:32,880 --> 00:05:37,520
ways too it can give us a much more

121
00:05:35,039 --> 00:05:39,440
thorough understanding of how organisms

122
00:05:37,520 --> 00:05:42,240
are related this can be useful for all

123
00:05:39,440 --> 00:05:44,039
sorts of things from veterinary sciences

124
00:05:42,240 --> 00:05:46,639
to rearing and taming to medical

125
00:05:44,039 --> 00:05:47,720
research to not naming animals pandas

126
00:05:46,639 --> 00:05:50,400
that aren't

127
00:05:47,720 --> 00:05:52,680
pandas it can also help us understand

128
00:05:50,400 --> 00:05:54,880
the whole evolutionary history of an

129
00:05:52,680 --> 00:05:57,400
organism if we know when it diverged

130
00:05:54,880 --> 00:05:59,319
from other organisms we can infer when

131
00:05:57,400 --> 00:06:01,600
certain traits must have evolved and

132
00:05:59,319 --> 00:06:03,680
that can help us Place extinct animals

133
00:06:01,600 --> 00:06:05,840
or any other animals that we don't have

134
00:06:03,680 --> 00:06:09,479
the DNA

135
00:06:05,840 --> 00:06:11,599
for so what was I trying to do I wanted

136
00:06:09,479 --> 00:06:14,599
to make a program that would allow me to

137
00:06:11,599 --> 00:06:16,039
explore a philogenetic tree of life also

138
00:06:14,599 --> 00:06:18,479
called a

139
00:06:16,039 --> 00:06:20,039
cladogram I thought that assuming a

140
00:06:18,479 --> 00:06:22,520
reasonable data source somewhere I

141
00:06:20,039 --> 00:06:26,080
should be able to make an interface that

142
00:06:22,520 --> 00:06:28,120
let you click around and see what the

143
00:06:26,080 --> 00:06:30,080
different relationships are see what

144
00:06:28,120 --> 00:06:32,960
organisms are in a clay find their

145
00:06:30,080 --> 00:06:35,360
closest relatives that kind of thing

146
00:06:32,960 --> 00:06:38,599
ultimately though my main goal my true

147
00:06:35,360 --> 00:06:41,960
desire was to be able to ask how closely

148
00:06:38,599 --> 00:06:44,080
related are these two things and what

149
00:06:41,960 --> 00:06:45,280
animals or what species I'm mostly

150
00:06:44,080 --> 00:06:48,720
interested in animals but it works for

151
00:06:45,280 --> 00:06:50,919
anything what species are in that clade

152
00:06:48,720 --> 00:06:54,599
specifically um remembering that a clade

153
00:06:50,919 --> 00:06:56,680
can both have clades and leaf nodes

154
00:06:54,599 --> 00:06:59,800
within it

155
00:06:56,680 --> 00:07:01,639
um so seeing the smallest clay that

156
00:06:59,800 --> 00:07:04,479
contained them both and what cles were

157
00:07:01,639 --> 00:07:06,319
in what species were in there from a

158
00:07:04,479 --> 00:07:09,080
computer science perspective this

159
00:07:06,319 --> 00:07:11,000
doesn't sound too hard ultimately this

160
00:07:09,080 --> 00:07:13,960
is just a tree like you've probably

161
00:07:11,000 --> 00:07:16,280
worked with before finding a sub tree is

162
00:07:13,960 --> 00:07:17,479
not too hard nor is finding a shared

163
00:07:16,280 --> 00:07:21,840
parent

164
00:07:17,479 --> 00:07:21,840
node assuming a reasonable data

165
00:07:25,599 --> 00:07:30,199
source so if this is so simple surely it

166
00:07:28,400 --> 00:07:33,240
exists already right

167
00:07:30,199 --> 00:07:36,199
well you'd think that wouldn't

168
00:07:33,240 --> 00:07:38,639
you but I hunted around and found a

169
00:07:36,199 --> 00:07:41,360
variety of tools that were not quite

170
00:07:38,639 --> 00:07:43,960
what I wanted I found ones that had many

171
00:07:41,360 --> 00:07:46,039
of these features but you needed to

172
00:07:43,960 --> 00:07:48,599
upload the data in a specific format

173
00:07:46,039 --> 00:07:50,520
that you already had I found ones that

174
00:07:48,599 --> 00:07:54,400
didn't have lots of the features I

175
00:07:50,520 --> 00:07:55,759
wanted and I found one that did almost

176
00:07:54,400 --> 00:07:58,639
everything I

177
00:07:55,759 --> 00:08:00,919
wanted uh it had a great interface it

178
00:07:58,639 --> 00:08:04,039
had all of life it had a good search

179
00:08:00,919 --> 00:08:07,159
function the only thing it didn't have

180
00:08:04,039 --> 00:08:09,000
was a way to give me uh was give me a

181
00:08:07,159 --> 00:08:11,479
way to ask how closely related are these

182
00:08:09,000 --> 00:08:14,400
two things and get all that

183
00:08:11,479 --> 00:08:16,759
information we'll come back to that one

184
00:08:14,400 --> 00:08:19,080
later so having established that nothing

185
00:08:16,759 --> 00:08:22,680
did quite what I wanted I thought to

186
00:08:19,080 --> 00:08:26,639
myself hey this sounds actually useful

187
00:08:22,680 --> 00:08:28,400
to real scientists though um surely

188
00:08:26,639 --> 00:08:30,520
somebody needs to at least occasionally

189
00:08:28,400 --> 00:08:33,719
say how closely related to these two

190
00:08:30,520 --> 00:08:35,159
things maybe I should ask a biologist so

191
00:08:33,719 --> 00:08:37,640
I reached out to the only practicing

192
00:08:35,159 --> 00:08:39,919
biologist I know my

193
00:08:37,640 --> 00:08:43,039
uncle here's a brief introduction to my

194
00:08:39,919 --> 00:08:44,600
uncle Dr Russell reichelt he has had

195
00:08:43,039 --> 00:08:47,600
many notable achievements in his

196
00:08:44,600 --> 00:08:49,120
lifetime and here's just a few of them

197
00:08:47,600 --> 00:08:52,160
he's a director of the Great Barrier

198
00:08:49,120 --> 00:08:54,519
Reef Foundation he is a board member of

199
00:08:52,160 --> 00:08:56,640
the Australian climate change Authority

200
00:08:54,519 --> 00:09:00,000
he is an officer of the order of

201
00:08:56,640 --> 00:09:01,399
Australia an allaround top-notch bloke

202
00:09:00,000 --> 00:09:04,959
and most importantly on his list of

203
00:09:01,399 --> 00:09:07,440
achievements he's my uncle so I wrote

204
00:09:04,959 --> 00:09:10,360
him an email and laid out the idea that

205
00:09:07,440 --> 00:09:12,040
I had the investigation that I had done

206
00:09:10,360 --> 00:09:14,640
a few questions and asked if he thought

207
00:09:12,040 --> 00:09:16,560
it was worthwhile and he basically said

208
00:09:14,640 --> 00:09:18,200
I don't do any of that kind of stuff

209
00:09:16,560 --> 00:09:21,959
anymore I mostly deal with policies and

210
00:09:18,200 --> 00:09:25,079
politics you need to talk to Dr Pat

211
00:09:21,959 --> 00:09:27,600
Hutchings Dr Pat Hutchings is a senior

212
00:09:25,079 --> 00:09:30,160
fellow at the Australian Museum and has

213
00:09:27,600 --> 00:09:33,240
done extensive research on the ology and

214
00:09:30,160 --> 00:09:34,920
philogyny of Australian polys which are

215
00:09:33,240 --> 00:09:37,160
cool little undersea

216
00:09:34,920 --> 00:09:39,640
worms so I had a great phone call with

217
00:09:37,160 --> 00:09:41,320
her and she gave me some good advice and

218
00:09:39,640 --> 00:09:44,720
shared with me the most unforgettable

219
00:09:41,320 --> 00:09:47,360
sentence anybody has ever said to me she

220
00:09:44,720 --> 00:09:49,279
said Tim flanner would be a good person

221
00:09:47,360 --> 00:09:52,880
to talk to about this but he's a bit

222
00:09:49,279 --> 00:09:54,240
busy so he might not have the time now

223
00:09:52,880 --> 00:09:57,120
for those of you who don't follow this

224
00:09:54,240 --> 00:10:00,560
sort of thing or are too young Tim flry

225
00:09:57,120 --> 00:10:02,519
was Australian of the year in 2007

226
00:10:00,560 --> 00:10:05,399
while he would no doubt be an excellent

227
00:10:02,519 --> 00:10:07,640
person to talk to about this he is also

228
00:10:05,399 --> 00:10:10,480
unfortunately a bit

229
00:10:07,640 --> 00:10:13,600
busy it was then I realized I had made a

230
00:10:10,480 --> 00:10:16,360
tactical blunder uh in my head what I

231
00:10:13,600 --> 00:10:18,440
had done was present my idea to the

232
00:10:16,360 --> 00:10:19,760
closest biologist at hand and ask hey is

233
00:10:18,440 --> 00:10:20,839
this a good idea can you answer some

234
00:10:19,760 --> 00:10:23,640
questions for

235
00:10:20,839 --> 00:10:25,880
me in reality what I had done was

236
00:10:23,640 --> 00:10:28,680
present my half-baked beginnings of an

237
00:10:25,880 --> 00:10:29,920
idea to an extremely distinguished

238
00:10:28,680 --> 00:10:32,160
scientist

239
00:10:29,920 --> 00:10:34,000
who had generously introduced me to

240
00:10:32,160 --> 00:10:36,800
another extremely distinguished

241
00:10:34,000 --> 00:10:39,120
scientist who had in all sincerity and

242
00:10:36,800 --> 00:10:40,920
with the utmost of kindness tried to

243
00:10:39,120 --> 00:10:43,240
find the best person to answer my

244
00:10:40,920 --> 00:10:46,040
questions and so she recommended the

245
00:10:43,240 --> 00:10:49,800
Australian of the Year from

246
00:10:46,040 --> 00:10:53,680
2007 I did not reach out to Tim

247
00:10:49,800 --> 00:10:56,800
Flannery instead I spoke to Dr jacn Nan

248
00:10:53,680 --> 00:10:59,360
who Dr Pat Hutchings also recommended uh

249
00:10:56,800 --> 00:11:01,399
she is a postdoctoral researcher at the

250
00:10:59,360 --> 00:11:04,120
Australian Museum and she is focused on

251
00:11:01,399 --> 00:11:07,120
the evolution of birds and she was

252
00:11:04,120 --> 00:11:11,079
recently involved in a project reworking

253
00:11:07,120 --> 00:11:14,720
the whole philogyny of all birds a very

254
00:11:11,079 --> 00:11:17,079
relevant project to exactly this kind of

255
00:11:14,720 --> 00:11:18,920
Investigation Jacqueline was a huge help

256
00:11:17,079 --> 00:11:20,880
she answered my questions she pointed me

257
00:11:18,920 --> 00:11:23,079
at some potential data sources and

258
00:11:20,880 --> 00:11:25,399
generally confirmed that being able to

259
00:11:23,079 --> 00:11:28,720
see how two species were related would

260
00:11:25,399 --> 00:11:30,839
be a good tool and so armed with her

261
00:11:28,720 --> 00:11:32,760
advice and some confidence that this was

262
00:11:30,839 --> 00:11:35,399
a good project I started diving into the

263
00:11:32,760 --> 00:11:37,160
meat of the project and pretty quickly

264
00:11:35,399 --> 00:11:40,040
hit some

265
00:11:37,160 --> 00:11:43,079
snags the first data source I looked at

266
00:11:40,040 --> 00:11:45,639
was I the integrated taxonomic

267
00:11:43,079 --> 00:11:48,120
information system and it had many

268
00:11:45,639 --> 00:11:49,920
positive features that looked helpful

269
00:11:48,120 --> 00:11:53,360
you could search by common name which

270
00:11:49,920 --> 00:11:55,440
seemed ideal for enabling casual use it

271
00:11:53,360 --> 00:11:58,120
had really clear documentation so it was

272
00:11:55,440 --> 00:11:59,959
easy to access their API and even had a

273
00:11:58,120 --> 00:12:02,279
whole database dump of available for

274
00:11:59,959 --> 00:12:04,920
preloading all the data however it did

275
00:12:02,279 --> 00:12:06,440
have a couple of problems the first

276
00:12:04,920 --> 00:12:10,519
problem was that searching by common

277
00:12:06,440 --> 00:12:12,880
name was tricky there are so many common

278
00:12:10,519 --> 00:12:17,079
names suppose you search for Hedgehog

279
00:12:12,880 --> 00:12:20,760
you probably meant this right or did you

280
00:12:17,079 --> 00:12:24,800
maybe mean this or

281
00:12:20,760 --> 00:12:28,160
this or one of the 154 other options

282
00:12:24,800 --> 00:12:30,519
that it presents when you ask for a

283
00:12:28,160 --> 00:12:33,120
hedgehog there was also no way to filter

284
00:12:30,519 --> 00:12:35,800
these results nor are they sorted by

285
00:12:33,120 --> 00:12:38,040
anything other than ID uh there was

286
00:12:35,800 --> 00:12:40,959
nothing to do except give the user every

287
00:12:38,040 --> 00:12:42,920
single option and let them pick uh and

288
00:12:40,959 --> 00:12:45,639
that's bad enough a hedgehog imagine

289
00:12:42,920 --> 00:12:49,920
trying to pick from the 355 different

290
00:12:45,639 --> 00:12:52,279
carp results or the 472 sharks or the

291
00:12:49,920 --> 00:12:54,760
over 1,000

292
00:12:52,279 --> 00:12:57,720
worms this relates to the second problem

293
00:12:54,760 --> 00:13:01,680
I had the documentation was very clear

294
00:12:57,720 --> 00:13:03,560
but the API was sadly not suppose you

295
00:13:01,680 --> 00:13:06,480
had found the species you wanted and you

296
00:13:03,560 --> 00:13:09,760
had its ID so then you query the API to

297
00:13:06,480 --> 00:13:12,079
get the parent clade for it what returns

298
00:13:09,760 --> 00:13:13,760
is the ID for that clade so then you

299
00:13:12,079 --> 00:13:16,600
have to do a new API call to get the

300
00:13:13,760 --> 00:13:19,680
actual data of that clade okay that's

301
00:13:16,600 --> 00:13:21,760
not too bad let's try the reverse you

302
00:13:19,680 --> 00:13:23,800
have a CLA and you want to find all of

303
00:13:21,760 --> 00:13:26,639
its child clades and

304
00:13:23,800 --> 00:13:29,720
species so you query to get the children

305
00:13:26,639 --> 00:13:33,800
only it Returns the whole sub tree

306
00:13:29,720 --> 00:13:35,800
underneath that c and also only the IDS

307
00:13:33,800 --> 00:13:38,279
so now you have to make an API call for

308
00:13:35,800 --> 00:13:41,560
every ID there and check if it is a

309
00:13:38,279 --> 00:13:43,880
direct child and then for those ones

310
00:13:41,560 --> 00:13:44,880
make another API call and get all of

311
00:13:43,880 --> 00:13:48,040
their

312
00:13:44,880 --> 00:13:50,360
details compare just to compare just two

313
00:13:48,040 --> 00:13:53,240
quite related species like say a seal

314
00:13:50,360 --> 00:13:55,880
and a ferret you have to make over 100

315
00:13:53,240 --> 00:13:58,560
API calls and it wasn't particularly

316
00:13:55,880 --> 00:13:59,920
fast each of them took about a second

317
00:13:58,560 --> 00:14:02,440
and that's after you've picked from the

318
00:13:59,920 --> 00:14:05,360
12 fets and 144

319
00:14:02,440 --> 00:14:08,000
SEALS or things that return When you

320
00:14:05,360 --> 00:14:09,600
search those things but that's no

321
00:14:08,000 --> 00:14:11,000
problem right there's a whole database

322
00:14:09,600 --> 00:14:14,000
dump so you don't actually have to make

323
00:14:11,000 --> 00:14:16,519
these API calls right you can just load

324
00:14:14,000 --> 00:14:18,720
and read the database

325
00:14:16,519 --> 00:14:21,120
right well that brings me to the third

326
00:14:18,720 --> 00:14:23,920
problem which is one less about the data

327
00:14:21,120 --> 00:14:25,340
source and more about me uh I found that

328
00:14:23,920 --> 00:14:27,480
4 days

329
00:14:25,340 --> 00:14:30,920
[Music]

330
00:14:27,480 --> 00:14:33,759
ago so let's try this second data source

331
00:14:30,920 --> 00:14:36,639
this is called the open Tree of Life OT

332
00:14:33,759 --> 00:14:39,240
this data source seemed great it was

333
00:14:36,639 --> 00:14:41,240
super fast and it could do batch

334
00:14:39,240 --> 00:14:44,680
operations so you say get me all of this

335
00:14:41,240 --> 00:14:47,079
stuff for all of these organisms and it

336
00:14:44,680 --> 00:14:49,440
could even be given two or more species

337
00:14:47,079 --> 00:14:52,000
and just give you their most recent

338
00:14:49,440 --> 00:14:54,839
common CA or representation of the sub

339
00:14:52,000 --> 00:14:58,199
tree that contained all of them that

340
00:14:54,839 --> 00:15:00,399
sounds fantastic right well uh it

341
00:14:58,199 --> 00:15:02,959
actually was

342
00:15:00,399 --> 00:15:05,800
mostly um it couldn't do any common

343
00:15:02,959 --> 00:15:08,040
names at all but I figured maybe this

344
00:15:05,800 --> 00:15:10,199
was fine the common names were tricky

345
00:15:08,040 --> 00:15:12,399
with I so maybe that was something I

346
00:15:10,199 --> 00:15:15,120
didn't need at least for the first

347
00:15:12,399 --> 00:15:17,720
version so with the data source decided

348
00:15:15,120 --> 00:15:19,399
upon uh it was time to get to rendering

349
00:15:17,720 --> 00:15:21,720
and representing that data so let's have

350
00:15:19,399 --> 00:15:24,199
a look at what I'm trying to draw and

351
00:15:21,720 --> 00:15:26,399
and see what I'm trying to do this is a

352
00:15:24,199 --> 00:15:28,040
clog of all of the different living

353
00:15:26,399 --> 00:15:29,600
bears you can see here we've got a

354
00:15:28,040 --> 00:15:31,920
pretty classic tree structure that

355
00:15:29,600 --> 00:15:34,120
you've probably all seen before we've

356
00:15:31,920 --> 00:15:36,680
got a clay at the root node there that's

357
00:15:34,120 --> 00:15:38,880
iday and then each branch shows us

358
00:15:36,680 --> 00:15:41,639
different clades until we get down to

359
00:15:38,880 --> 00:15:45,079
the different species not every CLA

360
00:15:41,639 --> 00:15:48,279
necessarily has a name

361
00:15:45,079 --> 00:15:51,279
um often if you dig deep enough people

362
00:15:48,279 --> 00:15:53,639
have given these names but they'll all

363
00:15:51,279 --> 00:15:56,120
be sort of extremely similar to ur day

364
00:15:53,639 --> 00:16:00,920
and Ur a and on on on and

365
00:15:56,120 --> 00:16:00,920
on you tend to stick to a few big ones

366
00:16:01,399 --> 00:16:06,839
um this cogram is from a YouTube channel

367
00:16:04,279 --> 00:16:09,040
called Clint's reptiles uh Clint is a

368
00:16:06,839 --> 00:16:12,319
biologist who does a lot of videos about

369
00:16:09,040 --> 00:16:13,600
philogyny not just for reptiles and is

370
00:16:12,319 --> 00:16:15,639
how I got interested in all of this to

371
00:16:13,600 --> 00:16:18,199
begin with so that was kind of what I

372
00:16:15,639 --> 00:16:20,079
was imagining making at least visually

373
00:16:18,199 --> 00:16:22,240
I'd like to have some interactivity but

374
00:16:20,079 --> 00:16:23,800
this is what it might look like let's

375
00:16:22,240 --> 00:16:27,720
take a quick look at a different style

376
00:16:23,800 --> 00:16:29,680
of representation though this is and

377
00:16:27,720 --> 00:16:30,680
isn't a clog

378
00:16:29,680 --> 00:16:33,360
but it's a different style of

379
00:16:30,680 --> 00:16:35,240
representing the same things and it's a

380
00:16:33,360 --> 00:16:37,040
big fractal display where you can zoom

381
00:16:35,240 --> 00:16:38,880
in on each branch and keep zooming in

382
00:16:37,040 --> 00:16:42,199
and in and in and see what's there at

383
00:16:38,880 --> 00:16:44,680
every level um so we'll zoom in to the

384
00:16:42,199 --> 00:16:47,079
Bears and compare that so this is the

385
00:16:44,680 --> 00:16:52,440
same data as the last one all the same

386
00:16:47,079 --> 00:16:54,600
species represented with uh on on this

387
00:16:52,440 --> 00:16:56,319
platform so this representation is from

388
00:16:54,600 --> 00:16:58,120
one Zoom this is the tool that I

389
00:16:56,319 --> 00:16:59,880
mentioned earlier that does almost

390
00:16:58,120 --> 00:17:01,959
everything I want

391
00:16:59,880 --> 00:17:03,160
um and I think it's a great tool it's

392
00:17:01,959 --> 00:17:04,600
really great for just looking around

393
00:17:03,160 --> 00:17:07,880
seeing what life there

394
00:17:04,600 --> 00:17:10,079
is these are two very different styles

395
00:17:07,880 --> 00:17:12,480
and while both are good sticking with

396
00:17:10,079 --> 00:17:14,880
the classic cogram seemed like a simpler

397
00:17:12,480 --> 00:17:17,520
place to start so I started by using a

398
00:17:14,880 --> 00:17:19,799
library called biopython and they have a

399
00:17:17,520 --> 00:17:22,520
module called Pho which is explicitly

400
00:17:19,799 --> 00:17:24,160
for Pho genetics okay let's generate

401
00:17:22,520 --> 00:17:27,839
that same Bear

402
00:17:24,160 --> 00:17:29,600
tree oh I guess it draws it an asky

403
00:17:27,839 --> 00:17:31,360
that's super cool actually really like

404
00:17:29,600 --> 00:17:33,919
that but it's not quite what I want so

405
00:17:31,360 --> 00:17:36,000
let's try it again all right it's not

406
00:17:33,919 --> 00:17:38,960
asky anymore but there are two problems

407
00:17:36,000 --> 00:17:41,240
with this firstly this is way more bears

408
00:17:38,960 --> 00:17:44,760
right there were eight before and now

409
00:17:41,240 --> 00:17:48,080
there's what like 40 um so what

410
00:17:44,760 --> 00:17:50,160
gives well partly this is because open

411
00:17:48,080 --> 00:17:52,919
Tree of Life includes

412
00:17:50,160 --> 00:17:55,799
subspecies uh you can see wherever it's

413
00:17:52,919 --> 00:17:58,600
got three names that's a subspecies like

414
00:17:55,799 --> 00:18:01,039
Urus arctos horis up here that's the

415
00:17:58,600 --> 00:18:03,760
Griz bear which is a subspecies of brown

416
00:18:01,039 --> 00:18:08,360
bear it's also got extinct species like

417
00:18:03,760 --> 00:18:10,200
Urus ricus the pine Cave Bear up there

418
00:18:08,360 --> 00:18:12,600
now including subspecies and extinct

419
00:18:10,200 --> 00:18:15,080
species is kind of cool I'm not actually

420
00:18:12,600 --> 00:18:16,880
opposed to it um though maybe it should

421
00:18:15,080 --> 00:18:19,480
be a toggle because it does fill things

422
00:18:16,880 --> 00:18:22,000
up quite a bit there's a bigger problem

423
00:18:19,480 --> 00:18:26,000
though um this is just a

424
00:18:22,000 --> 00:18:29,440
PNG so biopython uses mat plop lib to

425
00:18:26,000 --> 00:18:31,720
render its phenetic trees and that just

426
00:18:29,440 --> 00:18:33,760
makes images which is great for putting

427
00:18:31,720 --> 00:18:36,039
in a research paper or putting on a

428
00:18:33,760 --> 00:18:38,159
website or something but at this point

429
00:18:36,039 --> 00:18:41,600
in the project we're getting to crunch

430
00:18:38,159 --> 00:18:42,960
time uh it's about 3 weeks out from Pon

431
00:18:41,600 --> 00:18:44,640
and I realized that if I'm going to make

432
00:18:42,960 --> 00:18:47,480
something that lets you actually explore

433
00:18:44,640 --> 00:18:50,200
these trees it's going to need something

434
00:18:47,480 --> 00:18:51,200
a little bit more substantial like an

435
00:18:50,200 --> 00:18:55,360
actual

436
00:18:51,200 --> 00:18:56,440
gooey so cure a bit of panic okay a lot

437
00:18:55,360 --> 00:18:59,559
of

438
00:18:56,440 --> 00:19:01,400
panic I'm now certain that I can't make

439
00:18:59,559 --> 00:19:04,200
anything even remotely like what I

440
00:19:01,400 --> 00:19:06,400
wanted to do before Pon uh and then I

441
00:19:04,200 --> 00:19:09,080
made another blunder I went and I looked

442
00:19:06,400 --> 00:19:10,600
at one Zoom again see when I had done

443
00:19:09,080 --> 00:19:11,919
all of the research and talking to the

444
00:19:10,600 --> 00:19:13,400
scientists and everything that was

445
00:19:11,919 --> 00:19:15,799
actually way back at the start of the

446
00:19:13,400 --> 00:19:17,960
year uh then this project sat on a shelf

447
00:19:15,799 --> 00:19:19,960
for a while while I moved house got

448
00:19:17,960 --> 00:19:23,799
married started a CTO and generally had

449
00:19:19,960 --> 00:19:25,600
an extremely busy year um remember when

450
00:19:23,799 --> 00:19:28,960
I showed you this slide before did you

451
00:19:25,600 --> 00:19:31,480
notice this button up here now now I

452
00:19:28,960 --> 00:19:33,080
don't know when that button was added uh

453
00:19:31,480 --> 00:19:35,440
but it either wasn't there when it f

454
00:19:33,080 --> 00:19:38,640
when I first looked or I didn't look

455
00:19:35,440 --> 00:19:39,840
very goodly Pro probably that one anyway

456
00:19:38,640 --> 00:19:44,679
look what happens when you click that

457
00:19:39,840 --> 00:19:47,320
button ooh two entry Fields so what

458
00:19:44,679 --> 00:19:51,039
happens if I enter Hedgehog and

459
00:19:47,320 --> 00:19:54,159
seal wow colored lines showing each

460
00:19:51,039 --> 00:19:57,559
lineage and it shows where they meet so

461
00:19:54,159 --> 00:19:59,360
let's zoom in on that and oh L theia the

462
00:19:57,559 --> 00:20:01,320
CL that specifically contains

463
00:19:59,360 --> 00:20:03,240
carnivorans and also

464
00:20:01,320 --> 00:20:04,039
hedgehogs yeah that sounds about right

465
00:20:03,240 --> 00:20:05,799
to

466
00:20:04,039 --> 00:20:08,960
me

467
00:20:05,799 --> 00:20:10,799
so where does that leave me what happens

468
00:20:08,960 --> 00:20:13,840
to my project now and what will I do

469
00:20:10,799 --> 00:20:17,880
with it thanks for coming back on talk

470
00:20:13,840 --> 00:20:20,960
no um I'll be serious truthfully I don't

471
00:20:17,880 --> 00:20:23,120
know um in some senses it might be worth

472
00:20:20,960 --> 00:20:25,159
continuing just because one Zoom is

473
00:20:23,120 --> 00:20:28,120
really good doesn't mean I couldn't also

474
00:20:25,159 --> 00:20:29,840
make something good um or I could try

475
00:20:28,120 --> 00:20:32,440
and contribute to one Zoom which is open

476
00:20:29,840 --> 00:20:35,400
source or maybe I'll just focus on that

477
00:20:32,440 --> 00:20:37,559
startup that I'm CTO of but I can't just

478
00:20:35,400 --> 00:20:39,919
leave without showing the answers to the

479
00:20:37,559 --> 00:20:42,000
questions I started with in the

480
00:20:39,919 --> 00:20:44,640
abstract so which animals are more

481
00:20:42,000 --> 00:20:46,679
closely related to cows than pigs

482
00:20:44,640 --> 00:20:49,880
everything along this red line which

483
00:20:46,679 --> 00:20:53,960
includes deer giraffes and

484
00:20:49,880 --> 00:20:56,640
whales are birds reptiles they sure are

485
00:20:53,960 --> 00:20:59,360
if they're not then neither is anything

486
00:20:56,640 --> 00:21:01,360
along the blue line but uh not the red

487
00:20:59,360 --> 00:21:02,760
line so that includes Turtles crocodiles

488
00:21:01,360 --> 00:21:04,440
and dinosaurs if you're willing to say

489
00:21:02,760 --> 00:21:06,799
none of those are reptiles you can say

490
00:21:04,440 --> 00:21:09,919
the birds aren't reptiles but if any of

491
00:21:06,799 --> 00:21:14,039
those are reptiles birds have to be as

492
00:21:09,919 --> 00:21:17,279
well am I a fish um yeah

493
00:21:14,039 --> 00:21:20,440
probably uh so there isn't a specific

494
00:21:17,279 --> 00:21:22,320
fish CL uh so what we can do is what's

495
00:21:20,440 --> 00:21:26,400
called philogenetic bracketing where you

496
00:21:22,320 --> 00:21:28,679
pick two different uh organisms that

497
00:21:26,400 --> 00:21:32,279
share the trait that you look that

498
00:21:28,679 --> 00:21:34,520
looking for and look at where they meet

499
00:21:32,279 --> 00:21:37,919
so in this case if we think of a reef

500
00:21:34,520 --> 00:21:39,919
fish like a clown fish you know Nemo um

501
00:21:37,919 --> 00:21:42,360
and then any cartilagenous fish like a

502
00:21:39,919 --> 00:21:45,679
shark or a ray or a

503
00:21:42,360 --> 00:21:49,559
sfish if both of those two things are

504
00:21:45,679 --> 00:21:53,600
fish uh then everything where those two

505
00:21:49,559 --> 00:21:57,240
lines meet at the Jord vertebrates uh

506
00:21:53,600 --> 00:22:00,640
must be fish as well um I have a jaw and

507
00:21:57,240 --> 00:22:03,320
a backbone uh I'm a Jord vertebrate and

508
00:22:00,640 --> 00:22:06,679
if you look way closely over there just

509
00:22:03,320 --> 00:22:09,640
in the top left it says

510
00:22:06,679 --> 00:22:13,000
mammals uh yep we're

511
00:22:09,640 --> 00:22:16,960
fish and lastly what is the last common

512
00:22:13,000 --> 00:22:19,320
ancestor of a bear and a weasel well

513
00:22:16,960 --> 00:22:21,840
actually it shows us a CLA but it

514
00:22:19,320 --> 00:22:25,159
doesn't show us any species within that

515
00:22:21,840 --> 00:22:28,159
CLA so I guess one Zoom doesn't do

516
00:22:25,159 --> 00:22:29,880
everything that I want it to do so maybe

517
00:22:28,159 --> 00:22:30,919
there's still some room for my project

518
00:22:29,880 --> 00:22:33,640
after

519
00:22:30,919 --> 00:22:36,320
all okay thanks for coming to my talk

520
00:22:33,640 --> 00:22:38,559
for real this time um I hope you learned

521
00:22:36,320 --> 00:22:40,960
how the best play best laid plans of

522
00:22:38,559 --> 00:22:43,080
fish and men can go wrong but maybe

523
00:22:40,960 --> 00:22:44,600
still be worth it I hope you learned

524
00:22:43,080 --> 00:22:46,960
something neat and interesting about

525
00:22:44,600 --> 00:22:48,840
Modern Biology and I hope that you're

526
00:22:46,960 --> 00:22:50,919
inspired to just try stuff out and then

527
00:22:48,840 --> 00:22:54,000
try talking to random experts and seeing

528
00:22:50,919 --> 00:22:55,760
how you go you never know and maybe I'll

529
00:22:54,000 --> 00:22:58,760
send a link of the recording of this to

530
00:22:55,760 --> 00:23:01,240
Tim Flannery

531
00:22:58,760 --> 00:23:01,240
thank you very

532
00:23:02,840 --> 00:23:09,840
[Applause]

533
00:23:06,200 --> 00:23:15,240
much thank you very much Jack for that

534
00:23:09,840 --> 00:23:15,240
uh delightful talk um do we have any

535
00:23:15,919 --> 00:23:21,559
questions we have a

536
00:23:19,320 --> 00:23:25,640
question can you say it in a big word

537
00:23:21,559 --> 00:23:29,200
big big voice this is Renee and me and

538
00:23:25,640 --> 00:23:30,440
Jack and a bear and you question for

539
00:23:29,200 --> 00:23:32,400
Jack rather than a comment can you

540
00:23:30,440 --> 00:23:35,080
phrase your question as a

541
00:23:32,400 --> 00:23:38,880
question

542
00:23:35,080 --> 00:23:42,120
yes why did pterodactyls fly what so the

543
00:23:38,880 --> 00:23:44,120
question was why did pterodactyls fly um

544
00:23:42,120 --> 00:23:47,440
a big reason why pterodactyls fly is

545
00:23:44,120 --> 00:23:49,080
because the sea levels were a lot higher

546
00:23:47,440 --> 00:23:51,279
um at the time when pterodactyl started

547
00:23:49,080 --> 00:23:53,320
to evolve and so instead of lots of big

548
00:23:51,279 --> 00:23:54,960
land like we have now there was lots of

549
00:23:53,320 --> 00:23:56,880
little islands and they needed to get

550
00:23:54,960 --> 00:24:00,159
between the islands and a really good

551
00:23:56,880 --> 00:24:01,440
way to get between islands is flying um

552
00:24:00,159 --> 00:24:04,520
and so a lot of the pterodactyls

553
00:24:01,440 --> 00:24:06,520
especially the really big ones uh they

554
00:24:04,520 --> 00:24:10,279
started flying because of

555
00:24:06,520 --> 00:24:10,279
that so to get between

556
00:24:11,520 --> 00:24:18,559
islands to get to the other

557
00:24:15,039 --> 00:24:20,279
side uh so you're a CTO that's pretty

558
00:24:18,559 --> 00:24:22,440
fancy a lot of things went wrong in this

559
00:24:20,279 --> 00:24:23,919
talk um well not in the talk the talk

560
00:24:22,440 --> 00:24:27,279
was great but the project had some

561
00:24:23,919 --> 00:24:28,919
hiccups how did you as a established

562
00:24:27,279 --> 00:24:30,720
developer come

563
00:24:28,919 --> 00:24:31,960
through those hurdles especially like

564
00:24:30,720 --> 00:24:33,760
from the perspective of someone who

565
00:24:31,960 --> 00:24:35,919
might like be a new developer how can

566
00:24:33,760 --> 00:24:37,880
they approach difficulties like that and

567
00:24:35,919 --> 00:24:40,799
keep going

568
00:24:37,880 --> 00:24:43,840
going yeah good question

569
00:24:40,799 --> 00:24:46,960
so I think it really depends on what

570
00:24:43,840 --> 00:24:49,840
you're trying to do

571
00:24:46,960 --> 00:24:52,360
if ultimately I was making this project

572
00:24:49,840 --> 00:24:55,240
because I was just messing around I I

573
00:24:52,360 --> 00:24:57,679
think that all this philogyny and and

574
00:24:55,240 --> 00:24:59,440
organisms and biology in general is

575
00:24:57,679 --> 00:25:02,480
super interesting and so I just like

576
00:24:59,440 --> 00:25:04,240
messing around with it um so I didn't

577
00:25:02,480 --> 00:25:06,240
necessarily need to keep going and when

578
00:25:04,240 --> 00:25:08,080
I hit snags I was more than capable of

579
00:25:06,240 --> 00:25:10,120
just dropping the project except for the

580
00:25:08,080 --> 00:25:11,600
fact that I had long since submitted to

581
00:25:10,120 --> 00:25:13,080
Pyon and got told that I was going to

582
00:25:11,600 --> 00:25:14,880
talk so I needed to come up with

583
00:25:13,080 --> 00:25:20,279
something

584
00:25:14,880 --> 00:25:24,960
um but if I was doing this for a uni

585
00:25:20,279 --> 00:25:28,840
assignment or for uh or my

586
00:25:24,960 --> 00:25:32,080
work I probably think about like taking

587
00:25:28,840 --> 00:25:33,600
a step back and asking myself you know

588
00:25:32,080 --> 00:25:36,320
what what are the problems I'm actually

589
00:25:33,600 --> 00:25:39,760
trying to solve how do

590
00:25:36,320 --> 00:25:42,000
I what am I getting stuck on is this

591
00:25:39,760 --> 00:25:43,240
something that I can put to the side and

592
00:25:42,000 --> 00:25:45,360
come back when I've learned more about

593
00:25:43,240 --> 00:25:47,480
the problem space is there something

594
00:25:45,360 --> 00:25:51,440
that I can actually just circumvent and

595
00:25:47,480 --> 00:25:55,039
solve other easier problems is there

596
00:25:51,440 --> 00:25:58,399
somebody on my team or in my general

597
00:25:55,039 --> 00:26:01,080
Network that might have an answer that I

598
00:25:58,399 --> 00:26:03,159
can talk to and get some advice from um

599
00:26:01,080 --> 00:26:05,760
talking to people is underrated and

600
00:26:03,159 --> 00:26:09,559
really good would

601
00:26:05,760 --> 00:26:11,360
recommend uh yeah would chat to Tim fler

602
00:26:09,559 --> 00:26:15,480
uh absolutely I would chat to Tim fler

603
00:26:11,360 --> 00:26:18,039
if he was not so busy thanks Jack sorry

604
00:26:15,480 --> 00:26:21,480
thanks Jack for the great talk um it

605
00:26:18,039 --> 00:26:23,320
inspires me to give a talk at Pon too um

606
00:26:21,480 --> 00:26:26,679
would you consider repurposing your

607
00:26:23,320 --> 00:26:30,000
software to find clades between

608
00:26:26,679 --> 00:26:32,000
languages um

609
00:26:30,000 --> 00:26:34,399
so I do think that that's quite

610
00:26:32,000 --> 00:26:36,600
interesting and I was uh when I watched

611
00:26:34,399 --> 00:26:40,480
the the languages in fact both of the

612
00:26:36,600 --> 00:26:43,360
previous talks in this room um I was

613
00:26:40,480 --> 00:26:45,919
thinking about the similarities and and

614
00:26:43,360 --> 00:26:48,200
the the the clustering the relationships

615
00:26:45,919 --> 00:26:50,799
between the languages and

616
00:26:48,200 --> 00:26:53,600
things I think that that could be very

617
00:26:50,799 --> 00:26:56,559
interesting I wonder how much of it

618
00:26:53,600 --> 00:26:59,320
would be sort of new and novel because

619
00:26:56,559 --> 00:27:00,679
relatively speaking there much fewer

620
00:26:59,320 --> 00:27:03,159
languages than

621
00:27:00,679 --> 00:27:06,240
organisms um and so

622
00:27:03,159 --> 00:27:09,480
maybe that clustering is like that

623
00:27:06,240 --> 00:27:11,159
categorizing already exists um but I'm

624
00:27:09,480 --> 00:27:14,399
also seeing a head shaking from somebody

625
00:27:11,159 --> 00:27:16,960
who knows things so maybe maybe it

626
00:27:14,399 --> 00:27:21,600
doesn't exist and it could be useful and

627
00:27:16,960 --> 00:27:21,600
interesting um so yeah maybe that's an

628
00:27:21,919 --> 00:27:26,559
Avenue hi uh was there anything else in

629
00:27:24,919 --> 00:27:28,320
your chat to the scientists that they

630
00:27:26,559 --> 00:27:32,279
came up with that they wanted to do that

631
00:27:28,320 --> 00:27:35,080
was not part of your plans at the time

632
00:27:32,279 --> 00:27:38,480
um

633
00:27:35,080 --> 00:27:39,320
so the the a couple of the big things

634
00:27:38,480 --> 00:27:41,640
were

635
00:27:39,320 --> 00:27:45,600
that

636
00:27:41,640 --> 00:27:49,159
um well here as a slight maybe not quite

637
00:27:45,600 --> 00:27:51,760
answered your question one thing that

638
00:27:49,159 --> 00:27:54,519
I've not seen in any other in many phog

639
00:27:51,760 --> 00:27:58,919
gentic representations but is a big

640
00:27:54,519 --> 00:28:01,600
problem is uh horizontal Gene sharing

641
00:27:58,919 --> 00:28:05,559
um this is not a monodirectional

642
00:28:01,600 --> 00:28:08,279
tree um you species can and do

643
00:28:05,559 --> 00:28:13,080
crossbreed in the wild even surprisingly

644
00:28:08,279 --> 00:28:15,039
distant ones um and uh when I first

645
00:28:13,080 --> 00:28:17,720
talked to my uncle about it he said that

646
00:28:15,039 --> 00:28:19,080
one of the big problems is like in the

647
00:28:17,720 --> 00:28:23,519
past sort of four or five years or

648
00:28:19,080 --> 00:28:26,720
something we found an instance of

649
00:28:23,519 --> 00:28:29,480
two you unrelated species of coral

650
00:28:26,720 --> 00:28:32,399
become sort of hybridizing and making a

651
00:28:29,480 --> 00:28:34,960
new species of coral and how do you fit

652
00:28:32,399 --> 00:28:36,519
that into there and then you say well

653
00:28:34,960 --> 00:28:40,799
okay what's the last common ancestor and

654
00:28:36,519 --> 00:28:42,320
you're like H that gets a bit tricky um

655
00:28:40,799 --> 00:28:45,399
so I think that kind of thing is

656
00:28:42,320 --> 00:28:48,880
something that uh isn't shown for

657
00:28:45,399 --> 00:28:51,640
example in one zoom and uh might be

658
00:28:48,880 --> 00:28:53,559
something worth exploring or adding if I

659
00:28:51,640 --> 00:28:55,200
continue on with this and make my own

660
00:28:53,559 --> 00:28:59,720
representation um that kind of

661
00:28:55,200 --> 00:28:59,720
hybridization events and things

662
00:29:00,200 --> 00:29:04,519
well if you would join with me please in

663
00:29:02,399 --> 00:29:07,919
thanking Jack one more time for that

664
00:29:04,519 --> 00:29:07,919
talk thank you very

665
00:29:08,519 --> 00:29:12,679
much thank you