0:00

Tired of ads barging into your favorite

0:02

news podcasts? Good news ad-free

0:05

listening on Amazon music is included with

0:07

your prime membership Just head

0:09

to amazon.com/ad-free news podcasts to catch

0:11

up on the latest episodes Without

0:14

the ads shows ad-free

0:16

for subscribers This

0:30

is the Naked Scientist Hello,

0:38

welcome to this week's Naked Scientists, the

0:40

program where we bring you the latest

0:43

breakthroughs in science, technology and medicine with

0:45

me Chris Smith Coming up did

0:47

the Black Death kill so many in Europe

0:49

that it left an indelible mark in the

0:51

population's DNA? Well new research suggests

0:54

that tasty theory might in fact be

0:56

a myth Also could smaller

0:58

wine measures in pubs help us to

1:00

tackle our unhealthy relationship with alcohol and

1:02

later on the Cambridge Engineers who are

1:04

trying to plug the gaps in the

1:06

UK's potholed roads with science

1:09

from Cambridge University's Institute of Continuing

1:11

Education This is the

1:13

Naked Scientists First

1:22

this week archaeologists and geneticists have been

1:24

looking at the ancient DNA of almost

1:26

300 people who lived

1:28

in Cambridgeshire before and after the

1:30

Black Death, the plague that repeatedly

1:32

rampaged across medieval England between

1:34

the 14th and through until the 17th centuries

1:37

They've been able to put to the test a theory

1:40

that's been around for a while now that says because

1:42

the plague was so vicious Carrying off

1:44

as it did maybe a third of the

1:46

population of Europe at the time it had

1:48

the effect of altering the genetics of the

1:50

population Selecting for genetically fit

1:52

survivors and weeding out the

1:54

genetically more vulnerable The results

1:56

though suggest that's not true

2:00

also offers intriguing new details about the

2:02

evolving social landscape at the time. Some

2:05

of the victims studied in the research were

2:07

buried in mass graves, one

2:09

of which was found under an earlier

2:11

entrance to one of Cambridge's oldest colleges.

2:14

I venture there on as it turned out the coldest

2:16

day of the year so far to

2:18

meet researcher Christiana Scheib and

2:20

Corpus Christi College's Fiona Gilson.

2:23

Well we're standing at what was the original

2:25

entrance to Corpus Christi College. The college was

2:28

founded in 1352, just in the wake of

2:30

the Black Death, the plague, and

2:32

the original entrance to the college ran

2:35

across this passageway here from St. Bennet

2:37

Street past St. Bennet's Church and into

2:39

what's now called Old Court. Indeed, this

2:41

does look a bit like the tradesman's entrance.

2:44

I can see why you moved it. It's

2:46

much more prestigious now, but what's special about

2:48

this entrance? Well, some years ago

2:50

we had to do some work on the building

2:52

of what is now the Taylor Library and so

2:54

we excavated part of this pathway and when we

2:57

did that the archaeologists found that

2:59

there were some skeletons down there that were buried

3:01

in a kind of an unusual way. Is

3:04

it relevant that right next door to where we're

3:06

standing there is also a church? It's

3:08

very relevant and this in fact is the

3:10

oldest church in Cambridge and there are lots

3:13

of burials here, but they were standard burials.

3:15

The ones that we're talking about today were

3:17

found stacked and not properly buried as individuals

3:19

and so that's an indication that they were

3:22

probably buried quickly. The thinking by archaeologists who've

3:24

worked on them was that they were probably

3:26

victims of the plague and I think that

3:29

some evidence came out to indicate that that

3:31

was the case. Standing next to

3:33

me is one of those researchers who worked

3:35

on this. Tell us who you

3:37

are and why you've come along today.

3:40

Hello, I'm Dr. Christiana Scheib. I'm a

3:42

fellow at St. John's College and a

3:44

group leader at the Department of Zoology

3:46

here in Cambridge. You had access

3:48

to some of the material that came out from under

3:50

our feet. What were you looking for? I

3:53

had access to four individuals plus

3:55

a fragment of cranium, so five

3:57

individuals probably, for

4:00

was one, the human genome. We wanted

4:02

to know more about medieval Cambridge, but

4:04

also I was looking for evidence of

4:06

plague. There's been this question about

4:08

the plague, which is people say that

4:10

because it killed so many people, it

4:13

had an effect on the genetics of the

4:15

people that are around today. It killed off

4:17

a vulnerable group, left a survivor group. Were

4:20

you able to get at any of those sorts of

4:22

questions with this? That has been

4:24

a long standing question sort of in the area,

4:27

in the field. You would expect that any time

4:29

you have a pandemic, you will have a pressure

4:31

on the population that is affected to adapt. With

4:33

the plague, we know that the mortality rate was

4:35

30 to 50, 60 percent. It

4:38

was very high. You would expect there would be a

4:40

huge impact on the genetics of

4:42

the people who survived. That was

4:44

what this project was about. Was

4:46

it just here you looked at, or did

4:48

you have material from other burials around the

4:51

town? The project itself comprised more than

4:54

10 sites across Cambridge. We were going

4:56

all the way back to the Neolithic.

4:58

The earliest individuals are from an early

5:00

Neolithic monumental burial in Trumpington. The

5:03

most recent are from the mid-19th century,

5:05

from Holy Trinity Church just a few

5:07

doors down. We wanted to

5:09

get an idea of the overall population structure

5:11

of Cambridge through time and how this might

5:13

have affected them. The benefit

5:16

of that huge time window is, of course,

5:18

that straddles the time when the plague circulated

5:20

in Europe, doesn't it? Because it had gone

5:22

by the 1800s and you predated it a

5:24

bit. I guess you can then ask if

5:26

there's a before and after effect in terms

5:28

of any impact on the genetics. Exactly.

5:31

What we were primarily looking at initially

5:33

was the potential genetic impact of the

5:35

Black Death or the second pandemic. That

5:37

started in 1347 and came in recurring

5:42

waves up until the 18th century.

5:44

How did you actually do it? We screened

5:46

a lot of dead people for both human

5:48

genome, what their immune systems looked like, as

5:50

well as what diseases they had and whether

5:53

or not they did have plague. We had

5:55

two sets, people who had plague and then

5:57

people who didn't have plague. have

6:00

plague and really for this study we ended up

6:02

not looking so much at the people who had

6:04

plague because we know that those people actually

6:07

died from plague and what we're interested in

6:09

are the people who were living before the

6:11

general population before the second pandemic or the

6:13

black death and then the people who were

6:15

definitely born afterwards so therefore they were children

6:17

of survivors. And

6:19

I guess you can then ask the question

6:21

are there any genes different between those

6:24

two groups because if there was some

6:26

effect of the plague killing off vulnerable

6:28

people or preserving people who were more

6:30

immune I suppose you'd expect to see

6:32

them more or less represented in those

6:34

populations. Exactly that's exactly what we

6:36

were looking for is there a difference

6:38

in the genetics of the population before

6:40

the black death and the people who

6:42

were born presumably of survivors. And

6:45

what did you find? We actually

6:47

found that there was no difference I mean there

6:49

were minor differences but nothing that you would call

6:51

statistically significant. So that sort

6:53

of blows out of the water this idea

6:55

that plague did have a big molding effect

6:57

on the genetics of Europe. The social

7:00

impact was huge and the impact on

7:02

the individual's lives would have been huge

7:04

but from a population standpoint

7:06

probably because we had

7:09

so much mobility afterwards

7:11

we don't see a strong

7:14

genetic impact at least

7:16

in Cambridgeshire and it could also be because

7:18

of the way that the plague kills. It's

7:21

not maybe targeting a specific gene or something

7:24

like that it's maybe a more

7:26

complicated route

7:28

to death if you will and so therefore there's

7:30

a lot of things being worked on and therefore

7:32

there's not one particular target it sort of muddies

7:35

the waters. What about the

7:37

social question as in because you were

7:39

looking at a range of individuals from

7:41

a range of different backgrounds did

7:43

you see that any particular groups were more

7:45

susceptible because one of the things we see

7:48

with disease we often say people who are

7:50

in poor conditions or poor health generally they

7:52

don't have good food don't have good living

7:54

conditions and more vulnerable people who were better

7:56

fed etc less so was the plague taking

7:59

no prisoners or did it follow that pattern?

8:01

People have long said plague is an indiscriminate

8:03

killer, and that seems to be reflected in

8:06

our data. It doesn't matter how rich you

8:08

are or how poor you are. You are

8:10

still susceptible to the plague. Now,

8:12

whether you survived, if you had good resources,

8:14

that might have helped you survive. And unfortunately,

8:16

with this kind of study, we're only looking

8:18

at the dead people that we

8:20

can see. So we're seeing people who we

8:22

know died of the plague or people who

8:24

we don't know whether they were ever exposed.

8:26

We can assume they were exposed, but perhaps

8:28

they never were. Maybe they moved to the

8:30

country and they managed to avoid it, which

8:32

did happen amongst people who could afford it.

8:35

Can you answer a question that's outside the scope of

8:37

your present paper, but one which has baffled people

8:39

for a long time, which is why plague just

8:41

disappeared? Yes, this is

8:44

a really interesting question, and lots of people

8:46

are working on it. It could have to

8:48

do with the vectors. So we believe that

8:50

it's carried in fleas on rats, and

8:53

perhaps something changed in the environment

8:55

or the way that the bacterium

8:57

was infecting the vectors. And maybe

9:00

improvements in sanitation helped eventually

9:03

for it to die out.

9:05

But it seems unlikely

9:07

that it went away simply because everybody

9:09

became immune. And

9:11

are there any more plague pits that you can dip

9:13

into in Cambridge, or have you run out of resources

9:15

now? I think there's plague pits

9:17

everywhere. In fact, my other work is on the

9:20

first pandemic, the earlier plague, which came to England,

9:22

that I've worked on as part of the After

9:24

the Plague Project, but it's also known as the

9:26

plague of Justinian. So during the early Anglo-Saxon period,

9:28

we also had a pandemic that came to England

9:30

and affected people in very much the same way.

9:32

And you're working on that now? I am, yes. You'll

9:35

have to tell us what you find. I

9:37

will, I'll keep you updated. And hopefully it'll be a warmer

9:39

day. Yes, I hope so, I'll publish in the

9:41

summer. Christina Scheiber,

9:43

and before that, Fiona Gilson.

9:46

Fascinating stuff, albeit on a very

9:48

cold day. Meanwhile, if you're

9:50

interested in the effects that diseases can have

9:52

on past groups of people, this month's edition

9:55

of one of our spin-off shows, that's Naked

9:57

Genetics. Look it up wherever you get your

9:59

podcasts. Takes a look at... the appearance of

10:02

multiple sclerosis across northern Europe. Totally

10:04

changing the subject. Now more and

10:06

more people are knocking alcohol on the

10:08

head in January in a bid to

10:10

give their bodies a welcome break following

10:13

the boozy excesses of Christmas. But

10:15

one other, and some argue better approach is

10:17

just to drink less alcohol in the first

10:19

place and behavioural scientists think they might have

10:22

found a way to help us to do

10:24

exactly that. Alayni Mantzari is a lecturer at

10:26

City University of London. If you

10:28

go into the pub and you ask for a glass

10:30

of wine and you're asked whether you want a small

10:32

one or a large one, and you might say large,

10:35

it's usually a 250ml portion. It's

10:39

a third of a standard bottle of wine. So

10:42

we wanted to see whether not

10:44

offering that and tapping the

10:46

size that you can get to 175, which

10:50

is what pubs call the medium size,

10:53

whether that would affect how much

10:55

people would drink. How did

10:57

you actually do this and do it meaningfully so

10:59

that it wasn't just a lab experiment? Because we

11:02

know that when we try and do lab experiments

11:04

on people, we always end up wondering whether or

11:06

not we've biased people because they know they're in

11:08

a study. This had been done

11:10

in the lab before and we wanted to

11:12

see what happens when you go out into

11:14

the real world. So we

11:17

asked 21 pub spars in

11:19

restaurant in England to

11:21

take away their larger serving size so

11:23

that the largest they offered was in

11:26

the 175. The

11:28

pubs took part in a 12-week study. So

11:30

the first four weeks, they just

11:33

went about their business as usual. So they

11:35

offered all the serving sizes wine. We didn't

11:37

make any changes. And we

11:39

recorded, well, actually they just shared

11:41

their sales reports. And then

11:43

for the next four weeks, we asked them

11:46

to remove that large size of wine by

11:48

the glass. And

11:50

again, they shared their sales reports and we

11:52

could see the amount of wine that they

11:54

sold during those four weeks. And

11:56

then the final four weeks of the study, we asked them to

11:59

reinstate the amount of wine introduce the large glass

12:01

of wine and again they shared their

12:03

sales report and we could see how

12:05

much wine they sold. And

12:08

we took into account various things

12:10

because we know that the weather

12:12

influences how much people drink, the

12:14

time of year and other major events

12:16

where people might drink more on a bank holiday

12:18

for example. So we took these factors

12:22

into consideration, into our analysis. Do

12:24

they order differently if you take that top

12:27

one away? Yeah, so what we

12:29

found was that when you take the large glass

12:31

away, on average

12:33

each venue sold around

12:35

420 milliliters less wine per day, so that's

12:38

about 7.6% less. And

12:44

just to illustrate that, one

12:46

and a half large glasses of wine. So

12:48

not tremendous amounts, but

12:50

it was an effect that would cause significance,

12:53

so it did make a difference. When

12:55

you look at the point of sale

12:57

data, because you're lucky you got the tour

12:59

receipts effectively from these venues, so you can

13:01

tell really what the behaviour may have switched

13:04

to, not an individual but you can tell

13:06

what the volumes are doing, could

13:08

you see that people were compensating? If you

13:10

give them less of the big

13:12

ones, do they just drink more of the small ones

13:14

to compensate? The interesting thing that we

13:16

found is that when you don't offer that

13:18

large glass, which is usually the 250, we

13:22

thought that people would just switch to the immediately

13:24

smaller one, which is the 175. But

13:27

actually what we found is people were

13:29

switching more to the 125, which is

13:32

the smallest one you can get. And

13:34

they weren't switching to other beverages like

13:36

beer that are not wine, and so

13:38

you're not missing cases because someone

13:41

went and had a pint of lager rather

13:43

than a big glass of wine. Yeah,

13:45

that's a really good point. So we looked

13:47

at beer and cider and we didn't

13:49

find that this had any

13:51

effect on those drinks. What

13:53

we weren't able to look at is other things

13:56

such as cocktails and spirits, but

13:59

we know That the or cider

14:01

and wine. Is what

14:03

most people order. and from the drinks

14:06

that most people order, we didn't really

14:08

find any switching around going on. Would

14:11

you think. The. Public health implications of

14:13

this: Oh, is this something that that

14:15

policymakers should now be actively think he

14:17

about deploying because we know the alcohol

14:19

is a top five killer isn't if

14:21

you look at the World Health Organization

14:24

seekers. Alcohol causes more disease

14:26

and more loss of life than

14:28

many many other things that we

14:30

regard as major threats. This

14:33

is just one study so for

14:35

policymakers to make a decision, they

14:37

would probably want to see more

14:39

research. On the topic. But

14:41

this study does suggest that.

14:44

They. Could potentially be some regulation

14:46

about. The larger sizes.

14:49

And. Maybe capping. Yet there are many

14:51

things to consider, such as the acceptability.

14:54

Of this by the public. And

14:56

the hospitality industry and the alcohol

14:58

industry as well. Is it

15:00

me or have glasses of wine.larger in

15:02

recent years that because I'm sure when

15:05

when I was he must a student

15:07

days wasn't that long ago but you

15:09

didn't see people walking around with what

15:11

looked like a bucket. When a now

15:13

you can get a half a bottle

15:15

of wine in some of these wine

15:18

glasses the your routine me seeing for

15:20

sale. So there's been. Pure things that

15:22

have happened to the axle. Wine

15:24

glasses have gone up the this their

15:26

size. So we did some research within

15:29

our group at Cambridge. And

15:31

we measured wine glasses over time and

15:33

we saw that they. Have increased.

15:35

Especially since the nineties, Almost.

15:38

Four hundred percent in some cases. and

15:40

there is research that show that if

15:43

you how. Large A lot wine

15:45

glasses you tend to pour. More And them. But

15:48

also the portion size that are

15:50

being offered has increased. So when

15:52

I was at Uni I don't

15:54

think you could ever get a

15:56

two hundred and fifty. Men:

15:58

The Leader Glass as. standard class.

16:01

The standard used to be the 1 to 5. It's

16:05

shifted what's considered normal towards

16:07

the bigger end. Aleniamentzari

16:10

there. Music

16:31

in the programme is sponsored by Epidemic

16:33

Sound, perfect music for audio and video

16:35

productions. You're listening to The Naked

16:37

Scientists with me, Chris Smith. The

16:40

UK's leading motoring organisations, that's the

16:42

RAC and the AA, have said

16:44

the country's potholed roads are largely

16:46

to blame for a huge surge

16:48

in vehicle breakdowns. Potholes

16:50

are usually formed when water freezes in cracks

16:53

in the road and then expands. Engineers

16:55

here in Cambridge are working on a

16:57

research project to do something about it.

17:00

One aspect of it is knowing

17:02

where the potholes are in the

17:05

first place. And, as it turns

17:07

out, the new generations of high-tech,

17:09

sensor-laden cars on out overcrowded roads

17:11

are already generating reams of data

17:13

about the road conditions. This,

17:15

the Cambridge team intend to mine and

17:18

then act on by deploying autonomous vehicles

17:20

that can go and patch up damage

17:22

almost before it even gets started. And

17:24

they're doing that using new self-healing road

17:26

surface materials. And also, they're embedding sensors

17:29

in the road to make the job

17:31

even easier. I went along

17:33

to meet the brains behind the

17:35

project, our pothole saviours, Professors Uranus

17:38

Brilakis and Abir Alta Bar.

17:41

The car industry has evolved dramatically over

17:43

the last 40 years and is

17:45

now able to give us, in

17:47

real time, an amazing wealth of

17:49

information. They can detect cracks, potholes,

17:51

fallen street signs, damage-laden markings. It's

17:54

just we have no way of taking that on board. Are

17:56

you saying modern cars, because they're endowed with all

17:58

these cameras and sensors? They're just

18:01

naturally, passively collecting all that

18:03

information anyway. Yes,

18:05

some of the car companies we work

18:07

with collect something like 50 gigabytes per

18:09

minute. They process that on board, come

18:12

up with these detections and only share those

18:14

detections in a statistical way. In other words,

18:16

at least three vehicles of the same type

18:18

need to go over that defect before this

18:21

is reported. Where does the data

18:23

go then? It's stored on the car. Does

18:26

it then get shared centrally with the manufacturer

18:28

then? Yes, it goes straight

18:30

to the manufacturer's cloud and then this is where

18:32

we come in. The plan is to have a

18:35

cloud system, a digital twin, if you want

18:37

to say, of the physical infrastructure that is

18:39

able to take all that information and start

18:41

making decisions. What does this change mean for

18:43

me? Do I need to go and right

18:46

now fix the problem or is it a

18:48

problem that I can wait for the next

18:50

maintenance schedule? And then

18:52

as a result of that, send that information

18:54

on to a maintenance vehicle or

18:57

to humans to make decisions. Does

18:59

this mean we can potentially get ahead of problems?

19:02

You don't wake up one day and the road

19:04

has a one meter across breach in it. That must

19:06

start as a very small breach that then gets worse

19:08

and worse and worse. So does this mean we can

19:11

potentially get upstream of some of these problems and fix

19:13

them before they even become a headache

19:15

for motorists? Imagine a

19:17

world where rather than us inspecting the roads

19:19

once a year or once every two years,

19:22

as soon as three vehicles drive on

19:24

top of a specific defect, it's already

19:26

detected and reported. That means

19:28

that patrols will never form again

19:31

any time in the future. We can fix them

19:33

well before they even materialize. And

19:36

who would do the fixing and how? That's

19:39

where our colleagues in the smart materials

19:41

area come in when we have effectively

19:43

autonomous vehicles that have the added function

19:46

of being in repair vehicles who

19:48

can come in then and address that problem.

19:51

Do they already exist these autonomous vehicles

19:53

though? They don't. Another

19:56

element of the project we're trying to create, we

19:59

have already a team of people on the

20:01

robotics front as well as the smart materials

20:03

and who's trying to come up with a

20:06

multi-purpose vehicle that doesn't just solve one problem

20:08

but also do it using smart materials that

20:10

are able to be much

20:12

more longer lasting and able to sense themselves

20:14

and give us information that would otherwise not

20:17

have. So there you are

20:19

Abia, you'll know where the problem is, how

20:21

big it is and how urgent it

20:23

is. Now it's over to you,

20:25

how are you going to fix it? Yes,

20:27

so we're going to make

20:29

use of the data to

20:31

understand the condition of the

20:33

road, decide where to perform

20:35

proactive maintenance, what damage

20:38

needs to be fixed and at

20:40

the same time make the road

20:42

smarter so that the road communicates

20:44

with the vehicle on its state

20:46

of health. Ah,

20:48

you're really delivering two solutions here then.

20:50

One is using the data we've just

20:52

been hearing about to optimise and get

20:54

ahead of problems as they're emerging but

20:57

then also make the road itself better

20:59

so that it's basically more

21:01

resilient and better at telling us when it's got a

21:03

problem going forward. Exactly, the

21:05

plan is that we can introduce additives

21:08

in the materials, they can tell us

21:10

where there's strain, where there's stress, where

21:12

there's watering grass but at

21:14

the same time we can embed some sensors

21:16

so they can again report back to

21:19

their state and condition, add

21:23

materials that will enhance the resilience

21:25

so they last longer, potentially

21:28

they could self-heal as well so we don't

21:30

even have to worry about inspecting the pavement.

21:33

Well let's think about those two things in

21:36

turn. The sensors first, what

21:38

are they, how will they work and what

21:40

sort of information will they deliver that we

21:43

can't just learn from the cameras on people's

21:45

cars already? They

21:47

will provide inside information

21:50

because our pavements are layered and there's

21:53

depth, the vehicles very

21:55

much provide data on the surface of

21:57

the road but many of the problems

22:00

underneath. So we can

22:02

actually by embedding those sensors detect

22:04

potential problems even earlier. And

22:08

in terms of the materials, what's

22:11

the new materials that we could use

22:13

then that will make potholes a bit

22:15

less common? So we

22:17

can provide more resilient materials

22:20

so they are stronger, they

22:22

last longer, they bond more

22:24

with the substrate but

22:26

also provide materials that are

22:28

self-healing. So if a small

22:30

crack occurs they heal the damage themselves so

22:33

you don't even have to go and repair

22:35

them. How do they do that? A

22:37

simple solution is capsules

22:39

which contain a healing agent

22:41

in them. You embed those

22:43

in the pavement, a crack

22:46

will just propagate through these

22:48

capsules and rupture them and

22:50

the healing product comes out

22:52

and seals the crack. It

22:55

sounds amazing but this sounds

22:57

also expensive. Is this going to

22:59

be new roads going forward and

23:01

we put this into a road as we build a road

23:04

or is it retrofittable? Because it sounds to me like you're

23:06

going to have to dig roads up one way or the

23:08

other to get these new materials in but also to get

23:10

the sensors in. It will be both

23:12

of them so you can apply self-healing

23:14

repair materials. You can also

23:17

apply self-healing as a coating. For example

23:19

some cars have self-healing paint on them

23:22

so you don't have to dig the pavement

23:24

and embed those materials in it. They could

23:27

be applied to the surface. What

23:29

sort of a difference do you think this

23:31

will make ultimately? If we're able to implement

23:33

this how is this going to

23:35

turn things around? I

23:37

think we need to look at this as

23:40

a long-term and a short-term solution. On the

23:42

long-term front yes it will take us a

23:44

long time to replace all of our pavements

23:46

but the same thing happened with traffic lights

23:48

when we went from incandescent light bulbs to

23:51

LEDs. This didn't happen overnight we gradually replaced

23:53

them. Also in

23:55

terms of the roads themselves we don't have to

23:57

dig up the whole road sometimes it's just the

23:59

top layer. that could be resurfaced.

24:01

And resurfacing is something that happens fairly

24:03

frequently on many of our roads.

24:05

So it will be something that will take

24:08

many years, perhaps even a couple of decades,

24:10

but it is possible to update the whole

24:12

network. Roads in this country carry 70%

24:14

to 80% of

24:17

all goods and people. This

24:20

is not going to change in the future. This has

24:22

been stable for quite some time and at

24:25

the same time it's a very underfunded mode

24:27

of transport. It receives only 15% of the

24:29

public spending. So

24:32

unsurprisingly, without enough money and we hold

24:34

that traffic, we have problems and

24:37

we have to change that. UN is

24:39

Blackis, before him, Abir

24:41

Al Tabar. They'll be friends

24:43

with everyone in the UK before too long if

24:45

they get that to work. When

24:47

L2 SARS-CoV-2, the viral cause of

24:49

COVID-19, and we're going to take

24:51

a closer look at the stubborn

24:54

persistence of COVID-19 symptoms, this is

24:56

frequently called long COVID. And now

24:58

a new analysis of blood samples

25:00

from patients suggests that we might

25:02

be closer to understanding what leads

25:04

some people to develop long COVID,

25:07

while other people recover from the

25:09

infection without incident. Here's Professor

25:11

Honor Boyman, who is at the Department

25:13

of Immunology at the University of Zurich.

25:16

Long COVID is actually a

25:18

combination of different symptoms. While

25:22

we understand these symptoms,

25:24

we don't understand actually how

25:26

they are caused. Or

25:29

what keeps them active. We

25:31

said, let's measure a lot

25:33

of proteins in

25:36

the blood of patients that have

25:38

long COVID. And then

25:41

let's inquire what is

25:43

the difference between those that have long

25:45

COVID compared to those that

25:47

don't. Is the sort

25:49

of rationale behind this then that if you

25:52

look at enough people enough times and you

25:54

look at enough different chemicals, you can spot

25:56

patterns that keep coming up in people who

25:58

do have the problem. and are

26:00

not there in the people who don't have

26:03

the problem and that argues that those changes

26:05

must in some way be linked to the

26:07

problem. That is correct

26:09

and that's exactly what we

26:11

did. And then we asked

26:13

what is the most different

26:15

protein or protein group between

26:17

long COVID patients and those

26:19

that don't have long COVID.

26:22

And what is it? What

26:24

is the biggest difference? We

26:26

found that an arm of

26:28

the immune system is actually

26:31

overactive in patients

26:33

that have active long COVID.

26:36

It's actually a very

26:38

interesting part of the immune system.

26:40

It's also called the complement

26:43

system which usually

26:46

should be activated when we

26:48

have an acute virus infection

26:50

and should then go back

26:53

to its normal state. In

26:55

long COVID this return to the

26:57

normal state is not happening and

27:00

what this then pauses

27:02

is actually a state

27:04

where the body and

27:06

many of the body's cells,

27:09

not only immune cells but

27:11

normal cells, are in an

27:13

alarmed state. Were

27:16

you able to look upstream of

27:18

the COVID infections in these people

27:20

in order to reassure yourself that

27:22

it wasn't like that already and

27:25

that that makes people more susceptible

27:27

to getting long COVID versus long

27:29

COVID is this happening? That's

27:32

very true. So then the

27:34

question of course is how can

27:37

we test whether this

27:39

is a cause or a

27:41

consequence of their symptoms? So

27:44

what we then did is

27:46

compared COVID patients which

27:49

either did not have any long COVID

27:51

at all or got

27:54

long COVID but then recovered

27:56

versus those in whom long

27:58

COVID persisted. we could

28:00

see is a compliment system

28:03

recovered in those which

28:05

recovered from their symptoms and

28:08

always at the same time. If

28:11

it's staying high, the levels of

28:13

this is staying active all the time, something must

28:15

be driving it, it's like a switch has been

28:18

thrown. So what do you think is

28:20

doing that? How is that being achieved? Yes,

28:22

so initially we think it gets

28:25

activated by the virus but then

28:27

why doesn't it switch off? There

28:29

we think there are loops

28:33

and signals that keep it

28:35

active in patients that

28:37

suffer from long COVID. It's

28:40

almost like someone's turned the thermostat up and

28:42

the temperature set point is now higher so the

28:44

room ends up hotter. Is

28:46

there a way of turning that thermostat

28:48

back down? Can we detune the immune

28:50

response back down again to where it

28:52

should be in these people? So

28:55

that's of course a

28:57

very important question. Some

29:00

individuals seem to be able to turn

29:02

it down. So if

29:04

we take someone who had COVID

29:06

and then long COVID and spontaneously

29:10

recovers, this

29:12

compliment activation is normalized

29:15

and the symptoms go away. Why

29:18

that is the case? We don't know. On

29:20

the other hand, if somebody does

29:22

not recover and keeps

29:25

on having this compliment activation

29:27

turned on all the time,

29:31

then another possibility would

29:33

be to use specific

29:36

trucks that actually act

29:38

on tuning down a

29:40

compliment system. Given

29:43

that you've got what appears to be

29:45

a sort of molecular fingerprint for someone

29:47

with long COVID, could

29:49

this be the basis of a test then? Because

29:51

that's been the other thing, hasn't it, that's proved

29:53

a bit elusive, is helping people to get a

29:56

diagnosis, to account for their symptoms, give them the

29:58

reassurance that this is what's going on. going on

30:01

and then obviously manage their symptoms for them. So

30:03

can we use this as a test? This

30:06

is a possibility. We've identified

30:08

a central piece in the

30:10

puzzle to understanding Long COVID,

30:14

a mechanism that unifies

30:16

the different findings

30:19

that had been collected

30:21

previously. This can

30:23

of course help diagnose

30:25

Long COVID better and

30:28

hopefully in a second step it

30:30

might actually also provide new

30:33

targets for treatment.

30:35

On a boyman there in Zürich. Well

30:38

now it's time for our question of the

30:41

week and Will Tingle took a look at

30:43

this optical conundrum from listener Catherine who off

30:45

the back of cataract surgery was

30:47

wondering this about her new set of eyes.

30:51

Why is it with replacement

30:54

lenses which are fantastic? Does

30:56

facial recognition still work but

30:59

biometrics do not seem to? Good

31:02

question Catherine. To get to the bottom

31:04

of this unsightly problem here is Anglia

31:06

Ruskin University's Dr Nikita Thomas. Biometrics

31:09

are essentially measurements of biological

31:11

features and facial recognition technology

31:13

uses biometrics to measure certain

31:15

facial landmarks that generate a

31:18

unique identifier or sometimes it's

31:20

called a face print of

31:22

your own face and

31:24

the landmarks measured with this type of

31:26

technology include features related to your nose,

31:29

your ears, your chin and so on

31:31

but the landmarks related to the eyes

31:33

are the distance between the eyes and

31:36

the depth of the eye circuit. These

31:39

are measured by the software detecting where

31:41

your upper and lower eyelids are and

31:43

where the corners of your eyes are.

31:45

Interocular lens implants replace the lens in

31:48

your eye and that is actually located

31:50

inside of the eye so it's behind

31:52

the pupil and behind the iris which

31:55

is the coloured part of the eye.

31:57

As the facial recognition technology Only

32:00

looks at the eyelids and the corners

32:02

of the eye, which are structures outside

32:05

of the eye itself, it's unlikely that

32:07

the intraocular lens implants that

32:09

are inside of the eye would

32:11

affect the detection of these outside

32:13

structures. However, it is the

32:16

case that during the surgery your eyelids

32:18

are held open with an instrument called

32:20

a lid speculum. It is possible that

32:22

the surgery itself and the holding of

32:24

these eyelids may have slightly

32:27

altered the positions or the

32:29

elasticity of your eyelids post-surgery

32:32

so that the facial recognition software

32:34

may not detect your eyelids in

32:36

the exact same precise location as

32:39

they were pre-surgery. So, is there

32:41

anything that can be done to fix this issue? So

32:43

in this situation it would probably be

32:45

best to attempt to redo the facial

32:48

ID profile for both your phone and

32:50

the banking app. But if

32:52

the problem persists even after doing this, then

32:54

it points to more of a problem with

32:56

the facial recognition software itself. This

32:59

is especially true if the facial recognition

33:01

software is working on your phone but

33:03

not on the app, as technically they

33:05

should both be reading the same unique

33:08

profile. There you are. Thank you

33:10

so much to Anglia Ruskin University's Nikita

33:12

Thomas for the answer and to Catherine

33:14

for the question. Next time we're having

33:16

a go at this question sent in

33:18

from Miss Nacilia, who, still full of

33:20

Christmas spirit and presumably special Christmas bread,

33:22

had this culinary query about Stollen. Why

33:25

does Stollen go floppy in the toaster

33:27

instead of crispy? Thanks very

33:29

much to Will Tingle and do keep

33:31

your questions and answers coming in. Chris

33:33

at thenakedscientist.com. That's all we have

33:35

time for this week. Do join us on

33:37

Tuesday though when we're back with the final

33:40

instalment of our Titans of Science series. We'll

33:42

hear from a world-leading authority on Alzheimer's disease

33:44

and the genes behind it. That's Julie Williams.

33:46

Don't miss it. The

33:48

Naked Scientist comes to you from the University

33:50

of Cambridge's Institute of Continuing Education. It's supported

33:53

by Rolls-Royce. I'm Chris Smith. Thank you for

33:55

listening and until next time. Goodbye. Tired

34:18

of ads barging into your favorite news

34:20

podcasts? Good news. Ad-free

34:23

listening on Amazon Music is included with

34:25

your Prime membership. Just head

34:28

to amazon.com/ad-free news podcast to catch

34:30

up on the latest episodes without

34:32

the ads.