ABOUT THE SPEAKER
Joe DeRisi - Biochemist
Joe DeRisi hunts for the genes that make us sick. At his lab, he works to understand the genome of Plasmodium falciparum, the deadliest form of malaria.

Why you should listen

Joseph DeRisi is a molecular biologist and biochemist, on the hunt for the genomic basis of illness. His lab at UCSF is focused on the cause of malaria, and he's also poked into SARS, avian flu and other new diseases as they crop up. His approach combines scientific rigor with a nerd's boundary-breaking enthusiasm for new techniques -- one of the qualities that helped him win a MacArthur "genius" grant in 2004. A self-confessed computer geek, DeRisi designed and programmed a groundbreaking tool for finding (and fighting) viruses -- the ViroChip, a DNA microarray that test for the presence of all known viruses in one step.

In 2008, DeRisi won the Heinz Award for Technology, the Economy and Employment.

More profile about the speaker
Joe DeRisi | Speaker | TED.com
TED2006

Joe DeRisi: Solving medical mysteries

Filmed:
474,538 views

Biochemist Joe DeRisi talks about amazing new ways to diagnose viruses (and treat the illnesses they cause) using DNA. His work may help us understand malaria, SARS, avian flu -- and the 60 percent of everyday viral infections that go undiagnosed.
- Biochemist
Joe DeRisi hunts for the genes that make us sick. At his lab, he works to understand the genome of Plasmodium falciparum, the deadliest form of malaria. Full bio

Double-click the English transcript below to play the video.

00:12
How can we investigate
0
0
3000
00:15
this flora of viruses that surround us, and aid medicine?
1
3000
5000
00:20
How can we turn our cumulative knowledge of virology
2
8000
4000
00:24
into a simple, hand-held, single diagnostic assay?
3
12000
4000
00:28
I want to turn everything we know right now about detecting viruses
4
16000
3000
00:31
and the spectrum of viruses that are out there
5
19000
2000
00:33
into, let's say, a small chip.
6
21000
3000
00:36
When we started thinking about this project --
7
24000
2000
00:38
how we would make a single diagnostic assay
8
26000
3000
00:41
to screen for all pathogens simultaneously --
9
29000
3000
00:44
well, there's some problems with this idea.
10
32000
2000
00:46
First of all, viruses are pretty complex,
11
34000
4000
00:50
but they're also evolving very fast.
12
38000
4000
00:54
This is a picornavirus.
13
42000
1000
00:55
Picornaviruses -- these are things that include
14
43000
2000
00:57
the common cold and polio, things like this.
15
45000
3000
01:00
You're looking at the outside shell of the virus,
16
48000
2000
01:02
and the yellow color here are those parts of the virus
17
50000
3000
01:05
that are evolving very, very fast,
18
53000
2000
01:07
and the blue parts are not evolving very fast.
19
55000
2000
01:09
When people think about making pan-viral detection reagents,
20
57000
3000
01:12
usually it's the fast-evolving problem that's an issue,
21
60000
4000
01:16
because how can we detect things if they're always changing?
22
64000
2000
01:18
But evolution is a balance:
23
66000
2000
01:20
where you have fast change, you also have ultra-conservation --
24
68000
4000
01:24
things that almost never change.
25
72000
2000
01:26
And so we looked into this a little more carefully,
26
74000
3000
01:29
and I'm going to show you data now.
27
77000
1000
01:30
This is just some stuff you can do on the computer from the desktop.
28
78000
3000
01:33
I took a bunch of these small picornaviruses,
29
81000
2000
01:35
like the common cold, like polio and so on,
30
83000
2000
01:37
and I just broke them down into small segments.
31
85000
4000
01:41
And so took this first example, which is called coxsackievirus,
32
89000
3000
01:44
and just break it into small windows.
33
92000
2000
01:46
And I'm coloring these small windows blue
34
94000
2000
01:48
if another virus shares an identical sequence in its genome
35
96000
5000
01:53
to that virus.
36
101000
1000
01:54
These sequences right up here --
37
102000
2000
01:56
which don't even code for protein, by the way --
38
104000
2000
01:58
are almost absolutely identical across all of these,
39
106000
3000
02:01
so I could use this sequence as a marker
40
109000
4000
02:05
to detect a wide spectrum of viruses,
41
113000
2000
02:07
without having to make something individual.
42
115000
3000
02:10
Now, over here there's great diversity:
43
118000
2000
02:12
that's where things are evolving fast.
44
120000
2000
02:14
Down here you can see slower evolution: less diversity.
45
122000
4000
02:18
Now, by the time we get out here to, let's say,
46
126000
2000
02:20
acute bee paralysis virus --
47
128000
2000
02:22
probably a bad one to have if you're a bee ---
48
130000
2000
02:24
this virus shares almost no similarity to coxsackievirus,
49
132000
5000
02:29
but I can guarantee you that the sequences that are most conserved
50
137000
4000
02:33
among these viruses on the right-hand of the screen
51
141000
2000
02:35
are in identical regions right up here.
52
143000
3000
02:38
And so we can encapsulate these regions of ultra-conservation
53
146000
3000
02:41
through evolution -- how these viruses evolved --
54
149000
3000
02:44
by just choosing DNA elements or RNA elements
55
152000
3000
02:47
in these regions to represent on our chip as detection reagents.
56
155000
4000
02:51
OK, so that's what we did, but how are we going to do that?
57
159000
3000
02:54
Well, for a long time, since I was in graduate school,
58
162000
2000
02:56
I've been messing around making DNA chips --
59
164000
3000
02:59
that is, printing DNA on glass.
60
167000
2000
03:01
And that's what you see here:
61
169000
1000
03:02
These little salt spots are just DNA tacked onto glass,
62
170000
3000
03:05
and so I can put thousands of these on our glass chip
63
173000
3000
03:08
and use them as a detection reagent.
64
176000
2000
03:10
We took our chip over to Hewlett-Packard
65
178000
2000
03:12
and used their atomic force microscope on one of these spots,
66
180000
2000
03:14
and this is what you see:
67
182000
2000
03:16
you can actually see the strands of DNA lying flat on the glass here.
68
184000
3000
03:19
So, what we're doing is just printing DNA on glass --
69
187000
3000
03:22
little flat things -- and these are going to be markers for pathogens.
70
190000
4000
03:26
OK, I make little robots in lab to make these chips,
71
194000
3000
03:29
and I'm really big on disseminating technology.
72
197000
3000
03:32
If you've got enough money to buy just a Camry,
73
200000
3000
03:35
you can build one of these too,
74
203000
2000
03:37
and so we put a deep how-to guide on the Web, totally free,
75
205000
4000
03:41
with basically order-off-the-shelf parts.
76
209000
2000
03:43
You can build a DNA array machine in your garage.
77
211000
3000
03:46
Here's the section on the all-important emergency stop switch.
78
214000
3000
03:49
(Laughter)
79
217000
2000
03:51
Every important machine's got to have a big red button.
80
219000
3000
03:54
But really, it's pretty robust.
81
222000
2000
03:56
You can actually be making DNA chips in your garage
82
224000
3000
03:59
and decoding some genetic programs pretty rapidly. It's a lot of fun.
83
227000
4000
04:03
(Laughter)
84
231000
1000
04:04
And so what we did -- and this is a really cool project --
85
232000
4000
04:08
we just started by making a respiratory virus chip.
86
236000
2000
04:10
I talked about that --
87
238000
2000
04:12
you know, that situation where you go into the clinic
88
240000
2000
04:14
and you don't get diagnosed?
89
242000
2000
04:16
Well, we just put basically all the human respiratory viruses
90
244000
2000
04:18
on one chip, and we threw in herpes virus for good measure --
91
246000
3000
04:21
I mean, why not?
92
249000
1000
04:22
The first thing you do as a scientist is,
93
250000
2000
04:24
you make sure stuff works.
94
252000
1000
04:25
And so what we did is, we take tissue culture cells
95
253000
3000
04:28
and infect them with various viruses,
96
256000
2000
04:30
and we take the stuff and fluorescently label the nucleic acid,
97
258000
4000
04:34
the genetic material that comes out of these tissue culture cells --
98
262000
3000
04:37
mostly viral stuff -- and stick it on the array to see where it sticks.
99
265000
4000
04:41
Now, if the DNA sequences match, they'll stick together,
100
269000
2000
04:43
and so we can look at spots.
101
271000
2000
04:45
And if spots light up, we know there's a certain virus in there.
102
273000
2000
04:47
That's what one of these chips really looks like,
103
275000
2000
04:49
and these red spots are, in fact, signals coming from the virus.
104
277000
3000
04:52
And each spot represents a different family of virus
105
280000
3000
04:55
or species of virus.
106
283000
1000
04:56
And so, that's a hard way to look at things,
107
284000
2000
04:58
so I'm just going to encode things as a little barcode,
108
286000
2000
05:00
grouped by family, so you can see the results in a very intuitive way.
109
288000
4000
05:04
What we did is, we took tissue culture cells
110
292000
2000
05:06
and infected them with adenovirus,
111
294000
2000
05:08
and you can see this little yellow barcode next to adenovirus.
112
296000
4000
05:12
And, likewise, we infected them with parainfluenza-3 --
113
300000
3000
05:15
that's a paramyxovirus -- and you see a little barcode here.
114
303000
2000
05:17
And then we did respiratory syncytial virus.
115
305000
3000
05:20
That's the scourge of daycare centers everywhere --
116
308000
2000
05:22
it's like boogeremia, basically.
117
310000
2000
05:24
(Laughter)
118
312000
1000
05:25
You can see that this barcode is the same family,
119
313000
4000
05:29
but it's distinct from parainfluenza-3,
120
317000
2000
05:31
which gives you a very bad cold.
121
319000
2000
05:33
And so we're getting unique signatures, a fingerprint for each virus.
122
321000
3000
05:36
Polio and rhino: they're in the same family, very close to each other.
123
324000
3000
05:39
Rhino's the common cold, and you all know what polio is,
124
327000
2000
05:41
and you can see that these signatures are distinct.
125
329000
3000
05:44
And Kaposi's sarcoma-associated herpes virus
126
332000
3000
05:47
gives a nice signature down here.
127
335000
2000
05:49
And so it is not any one stripe or something
128
337000
2000
05:51
that tells me I have a virus of a particular type here;
129
339000
2000
05:53
it's the barcode that in bulk represents the whole thing.
130
341000
4000
05:57
All right, I can see a rhinovirus --
131
345000
2000
05:59
and here's the blow-up of the rhinovirus's little barcode --
132
347000
2000
06:01
but what about different rhinoviruses?
133
349000
2000
06:03
How do I know which rhinovirus I have?
134
351000
2000
06:05
There're 102 known variants of the common cold,
135
353000
3000
06:08
and there're only 102 because people got bored collecting them:
136
356000
3000
06:11
there are just new ones every year.
137
359000
2000
06:13
And so, here are four different rhinoviruses,
138
361000
2000
06:15
and you can see, even with your eye,
139
363000
2000
06:17
without any fancy computer pattern-matching
140
365000
2000
06:19
recognition software algorithms,
141
367000
2000
06:21
that you can distinguish each one of these barcodes from each other.
142
369000
3000
06:24
Now, this is kind of a cheap shot,
143
372000
2000
06:26
because I know what the genetic sequence of all these rhinoviruses is,
144
374000
3000
06:29
and I in fact designed the chip
145
377000
1000
06:30
expressly to be able to tell them apart,
146
378000
2000
06:32
but what about rhinoviruses that have never seen a genetic sequencer?
147
380000
4000
06:36
We don't know what the sequence is; just pull them out of the field.
148
384000
2000
06:38
So, here are four rhinoviruses
149
386000
2000
06:40
we never knew anything about --
150
388000
2000
06:42
no one's ever sequenced them -- and you can also see
151
390000
3000
06:45
that you get unique and distinguishable patterns.
152
393000
2000
06:47
You can imagine building up some library, whether real or virtual,
153
395000
3000
06:50
of fingerprints of essentially every virus.
154
398000
2000
06:52
But that's, again, shooting fish in a barrel, you know, right?
155
400000
3000
06:55
You have tissue culture cells. There are a ton of viruses.
156
403000
2000
06:57
What about real people?
157
405000
2000
06:59
You can't control real people, as you probably know.
158
407000
2000
07:01
You have no idea what someone's going to cough into a cup,
159
409000
4000
07:05
and it's probably really complex, right?
160
413000
3000
07:08
It could have lots of bacteria, it could have more than one virus,
161
416000
3000
07:11
and it certainly has host genetic material.
162
419000
2000
07:13
So how do we deal with this?
163
421000
1000
07:14
And how do we do the positive control here?
164
422000
2000
07:16
Well, it's pretty simple.
165
424000
2000
07:18
That's me, getting a nasal lavage.
166
426000
2000
07:20
And the idea is, let's experimentally inoculate people with virus.
167
428000
5000
07:25
This is all IRB-approved, by the way; they got paid.
168
433000
5000
07:30
And basically we experimentally inoculate people
169
438000
3000
07:33
with the common cold virus.
170
441000
1000
07:34
Or, even better, let's just take people
171
442000
2000
07:36
right out of the emergency room --
172
444000
1000
07:37
undefined, community-acquired respiratory tract infections.
173
445000
4000
07:41
You have no idea what walks in through the door.
174
449000
2000
07:43
So, let's start off with the positive control first,
175
451000
3000
07:46
where we know the person was healthy.
176
454000
2000
07:48
They got a shot of virus up the nose,
177
456000
2000
07:50
let's see what happens.
178
458000
1000
07:51
Day zero: nothing happening.
179
459000
2000
07:53
They're healthy; they're clean -- it's amazing.
180
461000
2000
07:55
Actually, we thought the nasal tract might be full of viruses
181
463000
2000
07:57
even when you're walking around healthy.
182
465000
1000
07:58
It's pretty clean. If you're healthy, you're pretty healthy.
183
466000
2000
08:00
Day two: we get a very robust rhinovirus pattern,
184
468000
4000
08:04
and it's very similar to what we get in the lab
185
472000
2000
08:06
doing our tissue culture experiment.
186
474000
1000
08:07
So that's great, but again, cheap shot, right?
187
475000
3000
08:10
We put a ton of virus up this guy's nose. So --
188
478000
2000
08:12
(Laughter)
189
480000
1000
08:13
-- I mean, we wanted it to work. He really had a cold.
190
481000
4000
08:17
So, how about the people who walk in off the street?
191
485000
4000
08:21
Here are two individuals represented by their anonymous ID codes.
192
489000
2000
08:23
They both have rhinoviruses; we've never seen this pattern in lab.
193
491000
4000
08:27
We sequenced part of their viruses;
194
495000
2000
08:29
they're new rhinoviruses no one's actually even seen.
195
497000
3000
08:32
Remember, our evolutionary-conserved sequences
196
500000
2000
08:34
we're using on this array allow us to detect
197
502000
2000
08:36
even novel or uncharacterized viruses,
198
504000
2000
08:38
because we pick what is conserved throughout evolution.
199
506000
4000
08:42
Here's another guy. You can play the diagnosis game yourself here.
200
510000
3000
08:45
These different blocks represent
201
513000
2000
08:47
the different viruses in this paramyxovirus family,
202
515000
2000
08:49
so you can kind of go down the blocks
203
517000
1000
08:50
and see where the signal is.
204
518000
2000
08:52
Well, doesn't have canine distemper; that's probably good.
205
520000
3000
08:55
(Laughter)
206
523000
2000
08:57
But by the time you get to block nine,
207
525000
2000
08:59
you see that respiratory syncytial virus.
208
527000
2000
09:01
Maybe they have kids. And then you can see, also,
209
529000
3000
09:04
the family member that's related: RSVB is showing up here.
210
532000
2000
09:06
So, that's great.
211
534000
1000
09:07
Here's another individual, sampled on two separate days --
212
535000
3000
09:10
repeat visits to the clinic.
213
538000
2000
09:12
This individual has parainfluenza-1,
214
540000
3000
09:15
and you can see that there's a little stripe over here
215
543000
2000
09:17
for Sendai virus: that's mouse parainfluenza.
216
545000
3000
09:20
The genetic relationships are very close there. That's a lot of fun.
217
548000
4000
09:24
So, we built out the chip.
218
552000
1000
09:25
We made a chip that has every known virus ever discovered on it.
219
553000
4000
09:29
Why not? Every plant virus, every insect virus, every marine virus.
220
557000
3000
09:32
Everything that we could get out of GenBank --
221
560000
2000
09:34
that is, the national repository of sequences.
222
562000
2000
09:36
Now we're using this chip. And what are we using it for?
223
564000
3000
09:39
Well, first of all, when you have a big chip like this,
224
567000
2000
09:41
you need a little bit more informatics,
225
569000
2000
09:43
so we designed the system to do automatic diagnosis.
226
571000
2000
09:45
And the idea is that we simply have virtual patterns,
227
573000
3000
09:48
because we're never going to get samples of every virus --
228
576000
2000
09:50
it would be virtually impossible. But we can get virtual patterns,
229
578000
3000
09:53
and compare them to our observed result --
230
581000
2000
09:55
which is a very complex mixture -- and come up with some sort of score
231
583000
4000
09:59
of how likely it is this is a rhinovirus or something.
232
587000
3000
10:02
And this is what this looks like.
233
590000
2000
10:04
If, for example, you used a cell culture
234
592000
2000
10:06
that's chronically infected with papilloma,
235
594000
2000
10:08
you get a little computer readout here,
236
596000
2000
10:10
and our algorithm says it's probably papilloma type 18.
237
598000
4000
10:14
And that is, in fact, what these particular cell cultures
238
602000
2000
10:16
are chronically infected with.
239
604000
2000
10:18
So let's do something a little bit harder.
240
606000
2000
10:20
We put the beeper in the clinic.
241
608000
1000
10:21
When somebody shows up, and the hospital doesn't know what to do
242
609000
3000
10:24
because they can't diagnose it, they call us.
243
612000
2000
10:26
That's the idea, and we're setting this up in the Bay Area.
244
614000
2000
10:28
And so, this case report happened three weeks ago.
245
616000
2000
10:30
We have a 28-year-old healthy woman, no travel history,
246
618000
3000
10:33
[unclear], doesn't smoke, doesn't drink.
247
621000
3000
10:36
10-day history of fevers, night sweats, bloody sputum --
248
624000
4000
10:40
she's coughing up blood -- muscle pain.
249
628000
2000
10:42
She went to the clinic, and they gave her antibiotics
250
630000
4000
10:46
and then sent her home.
251
634000
1000
10:47
She came back after ten days of fever, right? Still has the fever,
252
635000
4000
10:51
and she's hypoxic -- she doesn't have much oxygen in her lungs.
253
639000
3000
10:54
They did a CT scan.
254
642000
1000
10:55
A normal lung is all sort of dark and black here.
255
643000
4000
10:59
All this white stuff -- it's not good.
256
647000
2000
11:01
This sort of tree and bud formation indicates there's inflammation;
257
649000
3000
11:04
there's likely to be infection.
258
652000
2000
11:06
OK. So, the patient was treated then
259
654000
3000
11:09
with a third-generation cephalosporin antibiotic and doxycycline,
260
657000
4000
11:13
and on day three, it didn't help: she had progressed to acute failure.
261
661000
4000
11:17
They had to intubate her, so they put a tube down her throat
262
665000
3000
11:20
and they began to mechanically ventilate her.
263
668000
1000
11:21
She could no longer breathe for herself.
264
669000
2000
11:23
What to do next? Don't know.
265
671000
2000
11:25
Switch antibiotics: so they switched to another antibiotic,
266
673000
3000
11:28
Tamiflu.
267
676000
2000
11:30
It's not clear why they thought she had the flu,
268
678000
2000
11:32
but they switched to Tamiflu.
269
680000
2000
11:34
And on day six, they basically threw in the towel.
270
682000
2000
11:36
You do an open lung biopsy when you've got no other options.
271
684000
4000
11:40
There's an eight percent mortality rate with just doing this procedure,
272
688000
2000
11:42
and so basically -- and what do they learn from it?
273
690000
3000
11:45
You're looking at her open lung biopsy.
274
693000
2000
11:47
And I'm no pathologist, but you can't tell much from this.
275
695000
2000
11:49
All you can tell is, there's a lot of swelling: bronchiolitis.
276
697000
3000
11:52
It was "unrevealing": that's the pathologist's report.
277
700000
3000
11:55
And so, what did they test her for?
278
703000
3000
11:58
They have their own tests, of course,
279
706000
1000
11:59
and so they tested her for over 70 different assays,
280
707000
3000
12:02
for every sort of bacteria and fungus and viral assay
281
710000
3000
12:05
you can buy off the shelf:
282
713000
2000
12:07
SARS, metapneumovirus, HIV, RSV -- all these.
283
715000
3000
12:10
Everything came back negative, over 100,000 dollars worth of tests.
284
718000
4000
12:14
I mean, they went to the max for this woman.
285
722000
3000
12:17
And basically on hospital day eight, that's when they called us.
286
725000
3000
12:20
They gave us endotracheal aspirate --
287
728000
2000
12:22
you know, a little fluid from the throat,
288
730000
2000
12:24
from this tube that they got down there -- and they gave us this.
289
732000
2000
12:26
We put it on the chip; what do we see? Well, we saw parainfluenza-4.
290
734000
5000
12:31
Well, what the hell's parainfluenza-4?
291
739000
2000
12:33
No one tests for parainfluenza-4. No one cares about it.
292
741000
3000
12:36
In fact, it's not even really sequenced that much.
293
744000
3000
12:39
There's just a little bit of it sequenced.
294
747000
2000
12:41
There's almost no epidemiology or studies on it.
295
749000
2000
12:43
No one would even consider it,
296
751000
2000
12:45
because no one had a clue that it could cause respiratory failure.
297
753000
3000
12:48
And why is that? Just lore. There's no data --
298
756000
3000
12:51
no data to support whether it causes severe or mild disease.
299
759000
4000
12:55
Clearly, we have a case of a healthy person that's going down.
300
763000
3000
12:58
OK, that's one case report.
301
766000
3000
13:01
I'm going to tell you one last thing in the last two minutes
302
769000
2000
13:03
that's unpublished -- it's going to come out tomorrow --
303
771000
3000
13:06
and it's an interesting case of how you might use this chip
304
774000
3000
13:09
to find something new and open a new door.
305
777000
2000
13:11
Prostate cancer. I don't need to give you many statistics
306
779000
4000
13:15
about prostate cancer. Most of you already know it:
307
783000
3000
13:18
third leading cause of cancer deaths in the U.S.
308
786000
2000
13:20
Lots of risk factors,
309
788000
2000
13:22
but there is a genetic predisposition to prostate cancer.
310
790000
4000
13:26
For maybe about 10 percent of prostate cancer,
311
794000
2000
13:28
there are folks that are predisposed to it.
312
796000
2000
13:30
And the first gene that was mapped in association studies
313
798000
4000
13:34
for this, early-onset prostate cancer, was this gene called RNASEL.
314
802000
4000
13:38
What is that? It's an antiviral defense enzyme.
315
806000
3000
13:41
So, we're sitting around and thinking,
316
809000
2000
13:43
"Why would men who have the mutation --
317
811000
2000
13:45
a defect in an antiviral defense system -- get prostate cancer?
318
813000
5000
13:50
It doesn't make sense -- unless, maybe, there's a virus?"
319
818000
3000
13:53
So, we put tumors --- and now we have over 100 tumors -- on our array.
320
821000
6000
13:59
And we know who's got defects in RNASEL and who doesn't.
321
827000
3000
14:02
And I'm showing you the signal from the chip here,
322
830000
3000
14:05
and I'm showing you for the block of retroviral oligos.
323
833000
4000
14:09
And what I'm telling you here from the signal, is
324
837000
2000
14:11
that men who have a mutation in this antiviral defense enzyme,
325
839000
4000
14:15
and have a tumor, often have -- 40 percent of the time --
326
843000
4000
14:19
a signature which reveals a new retrovirus.
327
847000
4000
14:23
OK, that's pretty wild. What is it?
328
851000
3000
14:26
So, we clone the whole virus.
329
854000
1000
14:27
First of all, I'll tell you that a little automated prediction told us
330
855000
4000
14:31
it was very similar to a mouse virus.
331
859000
2000
14:33
But that doesn't tell us too much,
332
861000
1000
14:34
so we actually clone the whole thing.
333
862000
2000
14:36
And the viral genome I'm showing you right here?
334
864000
2000
14:38
It's a classic gamma retrovirus, but it's totally new;
335
866000
3000
14:41
no one's ever seen it before.
336
869000
1000
14:42
Its closest relative is, in fact, from mice,
337
870000
3000
14:45
and so we would call this a xenotropic retrovirus,
338
873000
4000
14:49
because it's infecting a species other than mice.
339
877000
3000
14:52
And this is a little phylogenetic tree
340
880000
2000
14:54
to see how it's related to other viruses.
341
882000
2000
14:56
We've done it for many patients now,
342
884000
3000
14:59
and we can say that they're all independent infections.
343
887000
3000
15:02
They all have the same virus,
344
890000
1000
15:03
but they're different enough that there's reason to believe
345
891000
3000
15:06
that they've been independently acquired.
346
894000
2000
15:08
Is it really in the tissue? And I'll end up with this: yes.
347
896000
2000
15:10
We take slices of these biopsies of tumor tissue
348
898000
3000
15:13
and use material to actually locate the virus,
349
901000
2000
15:15
and we find cells here with viral particles in them.
350
903000
4000
15:19
These guys really do have this virus.
351
907000
2000
15:21
Does this virus cause prostate cancer?
352
909000
2000
15:23
Nothing I'm saying here implies causality. I don't know.
353
911000
4000
15:27
Is it a link to oncogenesis? I don't know.
354
915000
2000
15:29
Is it the case that these guys are just more susceptible to viruses?
355
917000
4000
15:33
Could be. And it might have nothing to do with cancer.
356
921000
3000
15:36
But now it's a door.
357
924000
1000
15:37
We have a strong association between the presence of this virus
358
925000
3000
15:40
and a genetic mutation that's been linked to cancer.
359
928000
3000
15:43
That's where we're at.
360
931000
1000
15:44
So, it opens up more questions than it answers, I'm afraid,
361
932000
4000
15:48
but that's what, you know, science is really good at.
362
936000
2000
15:50
This was all done by folks in the lab --
363
938000
2000
15:52
I cannot take credit for most of this.
364
940000
1000
15:53
This is a collaboration between myself and Don.
365
941000
1000
15:54
This is the guy who started the project in my lab,
366
942000
3000
15:57
and this is the guy who's been doing prostate stuff.
367
945000
2000
15:59
Thank you very much. (Applause)
368
947000
3000

▲Back to top

ABOUT THE SPEAKER
Joe DeRisi - Biochemist
Joe DeRisi hunts for the genes that make us sick. At his lab, he works to understand the genome of Plasmodium falciparum, the deadliest form of malaria.

Why you should listen

Joseph DeRisi is a molecular biologist and biochemist, on the hunt for the genomic basis of illness. His lab at UCSF is focused on the cause of malaria, and he's also poked into SARS, avian flu and other new diseases as they crop up. His approach combines scientific rigor with a nerd's boundary-breaking enthusiasm for new techniques -- one of the qualities that helped him win a MacArthur "genius" grant in 2004. A self-confessed computer geek, DeRisi designed and programmed a groundbreaking tool for finding (and fighting) viruses -- the ViroChip, a DNA microarray that test for the presence of all known viruses in one step.

In 2008, DeRisi won the Heinz Award for Technology, the Economy and Employment.

More profile about the speaker
Joe DeRisi | Speaker | TED.com