Open-Protocols.srt

1
00:00:00,000 --> 00:00:02,640
Lenny Teytelman: I often start with one
of my favorite tweets from a

2
00:00:02,640 --> 00:00:08,700
postdoc at UC Riverside Who says
I'm looking, reading a paper,

3
00:00:08,940 --> 00:00:13,860
looking for a protocol in 1997.
Paper, as described the 96 finds

4
00:00:13,860 --> 00:00:19,380
96 paper as described an 87,
find 87 paper, it paywalled. And

5
00:00:19,380 --> 00:00:24,270
that's a very common and
frustrating experience. This is

6
00:00:24,270 --> 00:00:29,010
from a biologist, here's one
from a physicist, where he says

7
00:00:29,040 --> 00:00:32,400
devices were fabricated, as
previously described, previously

8
00:00:32,400 --> 00:00:36,090
described, previously described
and the original reference as

9
00:00:36,090 --> 00:00:39,840
devices were fabricated with
conventional methods, right? So

10
00:00:39,840 --> 00:00:43,620
good luck, while reading the
paper and figuring out exactly

11
00:00:43,620 --> 00:00:47,520
what was done and how to
replicate it and building it.

12
00:00:48,030 --> 00:00:53,340
And this is really the mission
of protocols.io is to

13
00:00:53,490 --> 00:00:57,240
improve what we publish to make
it easier to share our methods

14
00:00:57,300 --> 00:01:01,470
long before publishing and
preprint or semi privately. And

15
00:01:01,470 --> 00:01:04,830
then after you publish that, as
I said, with my own personal

16
00:01:04,830 --> 00:01:08,730
experience, to have a place
where we can come back to, to

17
00:01:08,730 --> 00:01:12,330
share corrections and
optimizations. I've shared the

18
00:01:12,330 --> 00:01:16,770
slides. So I don't want to do
much with the slides, I just

19
00:01:16,770 --> 00:01:21,390
want because this is really
quick to show a little bit of

20
00:01:21,390 --> 00:01:25,290
the platform and what I think
makes it unique. So our mission,

21
00:01:25,290 --> 00:01:30,120
as I said, is to make this
sharing easier and more rigorous

22
00:01:30,510 --> 00:01:35,760
protocols.io is open access
CC-BY license for everything

23
00:01:35,760 --> 00:01:39,330
that's public, if you're sharing
our business model is we charge

24
00:01:39,330 --> 00:01:42,810
for keeping things private,
private collaborations, if

25
00:01:42,810 --> 00:01:46,320
you're an industry and you're
never publishing, that's paid,

26
00:01:46,410 --> 00:01:50,100
if you're sharing publicly, just
as bioRxiv is free to read,

27
00:01:50,160 --> 00:01:54,210
free to share open access CC by.
And,

28
00:01:56,580 --> 00:01:59,850
you know, as I said, we started
in 2014. We're now at a point

29
00:01:59,850 --> 00:02:06,480
where we have over 9000 public
protocols. And here's one of my

30
00:02:06,480 --> 00:02:09,840
favorite examples. These are not
just PDFs, they're dynamic, and

31
00:02:09,840 --> 00:02:14,010
interactive. People add videos
inside to show you what exactly

32
00:02:14,010 --> 00:02:18,720
they mean grind the tissue, you
can show the 10 second video, we

33
00:02:18,720 --> 00:02:22,830
have catalogs from reagent
vendors, so you can detail

34
00:02:22,830 --> 00:02:27,000
exactly what do you use? What
are the materials that are

35
00:02:27,000 --> 00:02:31,620
needed, where you've ordered
that. And one of my favorite

36
00:02:31,620 --> 00:02:39,630
examples, is this protocol from
a researcher in Australia

37
00:02:39,630 --> 00:02:45,420
National University. And if you
look at this step in the

38
00:02:45,420 --> 00:02:48,210
protocol, doesn't matter what
the details are. But this step,

39
00:02:48,210 --> 00:02:51,870
think of it as a recipe for
cooking. This step says, do

40
00:02:51,870 --> 00:02:55,110
something for two to two minutes
timer, and it says seven

41
00:02:55,110 --> 00:02:59,610
minutes. And this is confusing.
And instead of an email

42
00:02:59,610 --> 00:03:04,410
exchange, or this being a final
paper, because protocols are

43
00:03:04,410 --> 00:03:07,710
dynamic and interactive, you can
mouse over on the step, you can

44
00:03:07,710 --> 00:03:10,170
click on it, you can ask a
question. And you can see on the

45
00:03:10,170 --> 00:03:14,310
right hand side, one of the
researchers reading it said wait

46
00:03:14,340 --> 00:03:18,120
to be two to seven minutes. What
does that mean? The question

47
00:03:18,120 --> 00:03:21,330
goes to the author and everyone
who's using the protocol. And

48
00:03:21,330 --> 00:03:24,510
you can see the reply right
there. So instead of private

49
00:03:24,510 --> 00:03:27,990
email conversations and
answering a question the same

50
00:03:27,990 --> 00:03:31,170
question 10 times to different
people, you sort of have a

51
00:03:31,170 --> 00:03:36,720
public FAQ, that being built up.
And with a click of a button,

52
00:03:37,440 --> 00:03:40,380
you can create a new version.
Again, going back to the first

53
00:03:40,410 --> 00:03:44,730
talk we heard today, you can
create a new version, that

54
00:03:44,730 --> 00:03:48,870
conversation is still there. But
this gets its DOI, you can

55
00:03:48,870 --> 00:03:54,270
cite specific versions. And now
you can see it says that step is

56
00:03:54,270 --> 00:03:57,630
fixed. There was a typo. The
author made version four and it

57
00:03:57,630 --> 00:04:01,050
says whoops, right, that should
be two to three minutes timer is

58
00:04:01,050 --> 00:04:06,300
not seven anymore. So it's
corrected. And if you go next to

59
00:04:06,300 --> 00:04:10,800
the title, there is a compare
button that allows you to see

60
00:04:11,250 --> 00:04:14,580
what is the same, what's
different between the steps,

61
00:04:14,610 --> 00:04:17,790
what was modified between the
versions, right? So you have a

62
00:04:17,790 --> 00:04:21,870
snapshot and a bird's eye view
of what are those differences?

63
00:04:21,870 --> 00:04:27,120
What are the tweaks and
optimizations? Now, obviously, I

64
00:04:27,120 --> 00:04:30,300
cannot change your protocol. I
cannot create a new version of

65
00:04:30,300 --> 00:04:35,190
it only the author, the owner of
the protocol can. But we do have

66
00:04:35,190 --> 00:04:38,640
just as on GitHub, we do have
copying and forking, so I can

67
00:04:38,640 --> 00:04:41,730
make a branch of your protocol.
You work with grasshoppers, I

68
00:04:41,730 --> 00:04:45,690
work with crickets, you're doing
this step at room temperature, I

69
00:04:45,690 --> 00:04:50,220
do it at 30 degrees. And so I
can make a clone of your

70
00:04:50,220 --> 00:04:54,540
protocol it becomes mine. I can
edit it as I use it with my

71
00:04:54,540 --> 00:04:58,380
equipment with my strain or
species. And then when I'm

72
00:04:58,380 --> 00:05:01,920
ready, I can share it and we try
it Give really clear credit to

73
00:05:01,920 --> 00:05:05,610
both the original authors and to
show what's the evolution of the

74
00:05:05,610 --> 00:05:11,520
protocol. So you can see on this
protocol itself, you can see

75
00:05:11,910 --> 00:05:15,570
that it is a fork from another
protocol. And you can look at

76
00:05:15,570 --> 00:05:19,260
that history, you can click on
the forks tab to see what's the

77
00:05:19,260 --> 00:05:22,680
evolution of the versions and
forks and the same compare and

78
00:05:22,680 --> 00:05:28,080
contrast functionality that
applies. There are a lot of it's

79
00:05:28,080 --> 00:05:31,620
a we have a really powerful
editor that is kind of like

80
00:05:31,620 --> 00:05:34,650
Microsoft Word or Google Docs,
concurrent editing, but really

81
00:05:34,650 --> 00:05:38,790
meant for methods with reagents,
which which data set did you

82
00:05:38,790 --> 00:05:42,030
use, right catalog numbers,

83
00:05:42,030 --> 00:05:47,010
videos, all of the things that I
mentioning. So check it out. It

84
00:05:47,010 --> 00:05:49,620
doesn't cost anything to create
an account start sharing

85
00:05:49,620 --> 00:05:52,890
protocols, as I said, is free.
You can always get in touch

86
00:05:52,890 --> 00:05:56,370
requests in one on one demo. For
more in depth, I don't have time

87
00:05:56,370 --> 00:06:00,810
to talk about everything we've
built since 2014. But for open

88
00:06:00,810 --> 00:06:03,930
science, specifically, I did
want to highlight some things.

89
00:06:03,930 --> 00:06:06,900
So we have communities. It's not
just a giant repository. We have

90
00:06:06,900 --> 00:06:10,110
9000 public protocols at this
point that's growing quickly.

91
00:06:10,650 --> 00:06:13,650
But we also let people create
communities where you can

92
00:06:13,680 --> 00:06:17,970
organize the methods together.
So this is this is a one year

93
00:06:17,970 --> 00:06:21,510
old, obviously, community of
people who are working on

94
00:06:21,510 --> 00:06:25,170
Coronavirus related methods.
There's half 1000. Researchers

95
00:06:25,170 --> 00:06:31,980
in this group 230 plus protocols
for sequencing testing for

96
00:06:32,430 --> 00:06:37,740
wastewater monitoring, tracking
variants. And because these are

97
00:06:37,740 --> 00:06:41,670
communities when new protocols
come in, or when updates happen,

98
00:06:42,330 --> 00:06:44,730
everyone is notified. Everyone
knows when there's a new

99
00:06:44,730 --> 00:06:47,280
version, a new fork a new
protocol added or when

100
00:06:47,280 --> 00:06:51,930
discussions happen that I showed
you. So it becomes a

101
00:06:51,960 --> 00:06:56,100
collaborative method development
workspace that's out in the

102
00:06:56,100 --> 00:06:59,100
open. You can also have private
side but I really wanted to

103
00:06:59,100 --> 00:07:02,970
focus on the mission and the
public side of this. And from

104
00:07:02,970 --> 00:07:06,210
these Coronavirus protocols I
really want to highlight because

105
00:07:06,210 --> 00:07:09,870
it's just an amazing example I
want to highlight this protocol

106
00:07:09,870 --> 00:07:13,710
from it's actually funded by
Wellcome Trust, the group that

107
00:07:13,710 --> 00:07:16,470
shared it from University of
Birmingham. This was the first

108
00:07:16,470 --> 00:07:21,120
protocol on Coronavirus that we
got on protocols.io

109
00:07:21,120 --> 00:07:28,230
January 25th of 2020. And what I
find amazing about it is that

110
00:07:28,230 --> 00:07:32,640
the group that shared it, you
can see it's actually a fork

111
00:07:32,640 --> 00:07:38,760
from there. This is a sequencing
of the Coronavirus RNA. This

112
00:07:38,760 --> 00:07:41,490
protocol is a fork from what
they shared in September of

113
00:07:41,490 --> 00:07:45,870
2019. Before we even had
Coronavirus. It's a fork from

114
00:07:45,870 --> 00:07:50,100
their Ebola protocol. And they
share the same 100 plus methods.

115
00:07:50,130 --> 00:07:54,390
This is funded by Wellcome Trust
and UKRI. Their work they

116
00:07:54,390 --> 00:07:59,370
shared this Ebola protocol in
September. And as soon as they

117
00:07:59,370 --> 00:08:04,020
got the SARS-COV2 to sequencing
going in January, with a click

118
00:08:04,020 --> 00:08:08,520
of a button they made a fork,
they changed those few things.

119
00:08:09,570 --> 00:08:13,260
And again, you can do the
Compare and contrast. And you

120
00:08:13,260 --> 00:08:17,010
can see what was in the Ebola
protocol. And it's tiny things

121
00:08:17,010 --> 00:08:25,140
right. As you can see, when you
click compare, the differences

122
00:08:25,260 --> 00:08:29,760
are really small. Let me
actually go back to the first

123
00:08:29,760 --> 00:08:32,520
version, it's easier to
demonstrate. So what they share

124
00:08:32,520 --> 00:08:38,520
on January 22. I said 25th it
was January 22. If you click the

125
00:08:38,520 --> 00:08:42,420
Compare to check it against the
bolo protocol, the differences

126
00:08:42,420 --> 00:08:47,610
are really tiny. They use the
same amount of dntps random

127
00:08:47,610 --> 00:08:52,410
hexamers. And you know step two
is the same step one, instead of

128
00:08:52,410 --> 00:08:58,650
10 microliters of Ebola RNA, you
need 11 microliters of SARS cov.

129
00:08:58,650 --> 00:09:01,260
To write it's a small
difference, but it's important.

130
00:09:01,830 --> 00:09:07,320
And this is really critical for
methods that microliter can make

131
00:09:07,320 --> 00:09:11,280
big difference for efficiency.
It's the same as my story with

132
00:09:11,280 --> 00:09:15,360
that one microliter to five
microliters making or breaking

133
00:09:15,360 --> 00:09:20,700
the protocol. And this is a
beautiful example because they

134
00:09:20,700 --> 00:09:22,920
were able to share it quickly.
They were already on the

135
00:09:22,920 --> 00:09:25,740
platform. They had the bullet
protocol out they shared it

136
00:09:25,740 --> 00:09:32,010
rapidly in a year. If you look
at this protocol, right? No,

137
00:09:32,010 --> 00:09:38,670
it's not a paper. But there are
120,000 views of it. There are

138
00:09:38,880 --> 00:09:46,020
2000 export into PDF printing it
out right 57 people made copies

139
00:09:46,020 --> 00:09:49,740
of it and modified it as they
use it in their research group

140
00:09:50,010 --> 00:09:54,270
to get the whole genome
sequencing source code 260

141
00:09:54,270 --> 00:09:55,590
comments of it.

142
00:09:56,730 --> 00:10:00,420
And no it's not a paper. We're
not a journal. It makes end up

143
00:10:00,420 --> 00:10:05,190
in a paper at some point. But
this is that rapid sharing

144
00:10:05,220 --> 00:10:11,580
preprint like sharing that is
critical to open science. And

145
00:10:11,580 --> 00:10:17,520
before I take questions, the
last thing that is really

146
00:10:17,520 --> 00:10:22,680
special for me in highlighting
is when you're not just sharing

147
00:10:22,680 --> 00:10:25,890
a supplementary, but you're
putting things in the right

148
00:10:25,890 --> 00:10:30,780
repository, same way that we put
protein sequences in PDB. And

149
00:10:30,780 --> 00:10:35,010
DNA sequences in GenBank. When
you put things in the right

150
00:10:35,010 --> 00:10:39,690
repository, they become
discoverable. And I showed the

151
00:10:39,690 --> 00:10:43,680
live demo so I skipped those
slides. And I love this example

152
00:10:43,680 --> 00:10:46,920
where a researcher in Chile is
asking does anybody have a

153
00:10:46,920 --> 00:10:50,850
protocol for RNA extraction from
primary cortical neuron

154
00:10:50,850 --> 00:10:55,440
cultures, a postdoc at UCSF says
there's a couple on protocols

155
00:10:55,440 --> 00:10:58,920
that I owe, but here's the one
that I recommend for you. And

156
00:10:59,280 --> 00:11:04,920
what is what feels so amazing to
me about this example is that

157
00:11:05,280 --> 00:11:10,500
when you look at the protocol
itself, that the postdoc pointed

158
00:11:10,500 --> 00:11:14,340
to that should work for this
Chilean scientists. The protocol

159
00:11:14,340 --> 00:11:16,770
accompanies a paper in the
journal Giga science on

160
00:11:16,770 --> 00:11:22,290
parasites of stickleback fish,
right. So it has nothing to do

161
00:11:22,350 --> 00:11:26,460
with cortical neuron cultures,
that paper is parasites or

162
00:11:26,460 --> 00:11:30,750
stickleback fish. But because
they didn't just add a PDF as a

163
00:11:30,750 --> 00:11:33,720
supplementary file to their
paper and chances that

164
00:11:33,720 --> 00:11:36,480
somebody's looking for protocol
for cortical neuron cultures

165
00:11:36,480 --> 00:11:40,410
would be reading this paper,
close to zero, but because the

166
00:11:40,410 --> 00:11:43,890
author's shared it openly in a
repository, not only did they

167
00:11:43,890 --> 00:11:47,010
make their paper more rigorous
and reproducible and easier for

168
00:11:47,010 --> 00:11:50,160
others to repeat, but they're
adding to the knowledge base

169
00:11:50,160 --> 00:11:54,480
where people can find their
protocol for good use it for

170
00:11:54,480 --> 00:11:57,420
their technique, use it for
their species with their

171
00:11:57,420 --> 00:12:02,520
equipment. And so I will end
there, there are a lot of

172
00:12:02,520 --> 00:12:05,430
funders that are now
recommending protocols.io

173
00:12:05,430 --> 00:12:09,930
over 500 journals. When you
submit a paper, I encourage you

174
00:12:09,930 --> 00:12:14,550
to put a protocol in protocols.io
and link to it. So take a

175
00:12:14,550 --> 00:12:18,150
look at the platform. And I'll
stop there and would be

176
00:12:18,150 --> 00:12:24,390
delighted to answer questions.
Boy, and the most. I'm rushing

177
00:12:24,390 --> 00:12:28,350
so much to make sure that
there's time for questions that

178
00:12:28,380 --> 00:12:31,890
the most important slide I
didn't put up. This is the team.

179
00:12:32,400 --> 00:12:35,520
I get the credit for the work I
didn't do I didn't program the

180
00:12:35,520 --> 00:12:39,510
apps. I didn't build the
platform. That's my These are my

181
00:12:39,510 --> 00:12:43,650
brilliant co founders Alexia and
Irina from day one of protocols

182
00:12:43,650 --> 00:12:46,500
that I owe Alexei has been
leading the engineering team,

183
00:12:46,890 --> 00:12:51,450
and we welcome feedback requests
suggestions, and we're working

184
00:12:51,450 --> 00:12:55,710
daily. Even though we launched
back in 2014. We're working

185
00:12:55,710 --> 00:13:01,140
daily on making it better and
making it work for researchers

186
00:13:01,140 --> 00:13:01,590
like you