Hello, and thank you for listening to the MicroBinFeed podcast. Here we will be
discussing topics in microbial bioinformatics. We hope that we can give you some
insights, tips, and tricks along the way. There's so much information we all
know from working in the field, but nobody writes it down. There is no manual,
and it's assumed you'll pick it up. We hope to fill in a few of these gaps. My
co-hosts are Dr. Nabil Ali Khan and Dr. Andrew Page. I am Dr. Lee Katz. Both
Andrew and Nabil work in the Quadram Institute in Norwich, UK, where they work
on microbes in food and the impact on human health. I work at Centers for
Disease Control and Prevention and am an adjunct member at the University of
Georgia in the U.S. Welcome to our next episode of MicroBinFeed podcast. Thank
you, Andrew, for letting me host this one, and we'll start with Andrew Page to
introduce himself to this panel. No, we'll start with who you are. Okay, so I
did my PhD from the University of Birmingham, and I moved to QIB in 2017. I
started working on antibiotic susceptibility mechanisms, and the last five years
I have studied different antibiotics, how the bugs become resistant to those
antibiotics, and what are the novel mechanisms, and what are the existing
mechanisms, and how they differ from the novel mechanisms. That's what my work
is about. So I have started in the last few months a science soundbite series
that helps scientists to understand scientific literature. So we try to make it
easier for the PhD students, postdocs, or other researchers, the work of other
scientists. And in today's discussion, we will be talking about mobile genetic
elements that are present in many bacteria, or almost all organisms, and we'll
be talking to our experts that are Dr. Andrew Page, Dr. Emma Waters, and Dr.
Heather Felgate. So Andrew Page, would you like to introduce yourself? Sure. I'm
Andrew Page, co-host of the MicroBioInfo podcast, and I'm head of informatics at
the Quadram Institute in Norwich in the UK. And I also do lots of informatics-y
stuff, and I've done a little bit of MG, mobile genetic elements stuff, but
really, you know, I've got the experts here in front of me who actually not only
do a bit of informatics, but also do a lot of work in the lab. So maybe Heather,
do you want to introduce yourself? Hi, so I'm Heather. I'm a postdoc in Mark
Webber's group looking at Staphylococci in general, carriage, and also isolates
that are implicated in infection in neonatal babies. So yeah, I've been working
here for quite a few years, but I've got a long history with Staphylococci and
microbiology in general. Thank you. So my name is Dr. Emma Waters. Previously, I
finished my PhD in 2017, and that was in chemistry. So there's a lot of biology
terms that I struggle with every now and again. But since then, I've done a
postdoc in the School of Chemistry at UVA, and then a couple of postdocs in
School of Medicine, and now I'm a postdoc in German Languages group at the
Quadram Institute. I focus on genome rearrangements in pretty much any bacteria,
but mostly I like Salmonella, definitely Typhi, and I do have a bit of
bioinformatics slash lab all-round experience. Almost all of us have commensal
Staphylococci or some species of Staphylococci. Do you know what is Salmonella
enterica serovar typhi is? It's a gram-negative bacteria that causes typhoid
fever. You might have heard of that, typhoid fever. It is quite common in
developing countries where sanitation and the sanitary conditions are not that
good, and water is contaminated. So Dr. Andrew Page worked with a group of
scientists from Pakistan in 2018, and they found out extremely drug-resistant
typhoid. So what is extremely drug-resistant typhoid? I can explain that first
before going on to that. So usually we treat Salmonella typhi, or the typhoid
fever, with three common antibiotics that are chloramphenicol, amoxicillin, and
cortamoxicillin. And if the typhoid, Salmonella typhi, become resistant to these
three antibiotics, that is called MDR typhi, multi-drug-resistant typhi. And
then the second line of drugs, if you want to say that, that is
chlorofluorquinolones, that like ciprofloxacin, and ceftriaxone, that is
cephalosporins. So if it becomes resistant to that one as well, then it's called
XDR. Then there are only one or two drugs left for other treatment options. And
they find out that in Pakistan, that strain is present, that is only susceptible
to azithromycin and carbapenems. That was a very dangerous condition. And
because of their work, not only the international activists wake up, but also
the local public health officers and machinery become activated. And just
because of that one publication, they had started quite a few multicenter
studies, and they had started a vaccine program now in Pakistan that has reduced
the incidence of typhoid in that southern part of the country. And now in the
northern part of the country, although they were trying to restrict it, but it
has been reported. And you know what? Dr. Emma Waters was the person who worked
with a group of other scientists from the northern part of Pakistan, and they
found out the transmission has happened in the northern part of the country, and
they reported it. And now they are trying to contain it there as well. So we'll
ask some questions to these two experts about how they identified XDR typhoid in
the first place in Pakistan. Would you like to tell us the story a little bit
about that? How did you find that? Yeah. So we got samples that were taken from
outbreaks in Pakistan, and obviously typhoid is a human pathogen, and it kills,
what, 100,000 people a year in Zaledama? So very, very bad, and it's usually the
people who are least able to access healthcare are the ones impacted, and it's
the fecal-oral route, if you know what that is. And so what we did was we got a
load of samples from the area which were not XDR, and a load of samples which
were. So you could tell from growing muffin plates and seeing the antibiotic
resistance. And then we used bioinformatics, and it meant having to write
bioinformatics tools to pull out what is similar, what is different, and out of
that we found, actually, there was a plasmid, basically, or we found evidence of
a plasmid broken into three pieces, and I was like, yeah, grand, bingo, there
you go, we've got a plasmid. It's introduced all of these extra AMR genes on top
of an already quite drug-resistant strain of typhoid, and that's not very good,
and then we could trace it back by looking in the public archives and see that
it came from an E. coli strain in cattle, which is picked up by CDC, kind of
global surveillance, and that's kind of a nice little story. You see it in one
totally different species, and it moves into a different one, and, yeah, and
that's kind of where we stopped. That was used as evidence for the WHO to roll
out a vaccination program, and so that, you know, that kicked off all the
machinery, as you said, Yasir. That's very interesting. So that work that you
did, was it solely based on sequencing, or was it based on some phenotypic
observations that they found in the bacteria? It was a combination of two, so it
was the phenotyping data plus the genome sequences, so we knew what they were.
All right. You found out that there are some genes that are present there, and
identified as they are salmonella genes, and then you found some additional
genes to that. So we can say that there are some core genomes of salmonella
there, present there, and then there are some accessory genes that came in.
Yeah, so the original strain was H58, which is like the original haplotyping
scheme that they had for typhi, and that's been causing havoc around the world,
you know, that's most of the multidrug-resistant cases of typhi, and obviously
that caused a lot of deaths, so it's the one that everyone really looks out for,
and then that acquired an extra plasmid, and, you know, plasmids move in and
out, and in this case, some of the drug-resistant genes had integrated into the
chromosome, so it's really bad, you know, if they're in a plasmid, they can move
out and be kicked out, but when they then get integrated, they're very difficult
to then get rid of, and then it got an extra plasmid on top of that, with extra
genes on top of that, and in some cases, they're duplicate gene, AMR genes, so
you've got, you know, twice as bad. So not good all around, and obviously this
happened because we overuse antibiotics and we need proper antibiotic
stewardship. Okay, that's pointed to a very good point, basically. So Emma,
would you like to tell us the story that how did you get involved in this?  that
finding the XDR typhoid in Pakistan and how did you manage to do that? So we
were contacted by a hospital in the north of the country and originally the
outbreak was in the south so they were aware of what was happening and then they
realised in this one hospital in Lahore that they were starting to get more and
more cases which weren't treatable with their normal antibiotic choice. So they
started to pick up an idea of what might be happening and that was when they
sent us samples to do sequencing of to basically see if the same thing was
happening, if that outbreak strain was going into the northern part of the
country. Then did you find any difference in the sequence of the organism that
is found in the southern part of the country and in the northern part of the
country or was it the same? So I think in this case it was the same so it was
the same sort of introduction into the system but correct me Andrew if I'm
wrong, it was the same introduction but on another occasion in a different
publication we actually have seen that this region has been introduced in
different ways so it's like the same resistance has been introduced in different
ways. So as we are talking about mobile genetic elements, Heather would you like
to introduce us what kind of mobile genetic elements are there in bacteria?
There's a lot, there's a lot of different mobile genetic elements and some can
jump between different species, some only like staying in one type of species,
it's all quite complicated and well that's why we have a job to I suppose look
for them. In regards to Staphylococcus it's very different to what Emma and you
guys work on with gram-negative salmonella and stuff so we just kind of have to
keep looking, keep searching and work out which ones are jumping from where to
here to there. So like there are phages and there are transposons and there are
plasmids but there are different kind of plasmids in gram-positive and gram-
negative, okay that makes sense. So how do we identify like there is one way of
sequencing it and then we can find there is a mobile genetic element but is
there any other way of finding it like through any phenotype mechanism or
through like any antibiotic susceptibility mechanism or we can identify that,
okay there is a plasmid there or usually it's easy just to do sequencing. So I
think they go hand in hand really, you have to do some sequencing and then you
have to check them phenotypically. Quite often I've found with Staphylococci
that it just takes them a while to wake up some days like they haven't had a
coffee so you know their genome might say they've got quacks so they might be
resistant to certain set of antibiotics but when you test them they're not and
it's like you can grow them in different ways to turn those genes on and off so
but you also have to look at the percentage of that gene, is it a full gene
that's there that's on the plasmid or has it you know suddenly got like been
truncated or snapped in half and so you've got to keep searching and keep like
practice how they look like on a plate and how they grow like in antibiotics and
also marry that up with their genotype because quite often I've found as well
that there might be genes that we don't know yet and they've got similar
antibiotic resistances but they don't have those genes. So you work at a lot of
genomes from hospitals and I hear that that's really really bad for antibiotic
resistance and so hospitals hospitals in general you know like they hand out
drugs like candy and of the samples that you know are antibiotic resistant how
many can you kind of link back what percentage can you link back to a genotype?
Staphylococci it's really difficult like I've been working on Staphylococci for
like four or five years or something now they're gonna kill me and there are
genes just unknown so I reckon about 80% of genes like when I see genes actually
then relate to a phenotype as well but quite as again quite often they might
have the gene and not be resistant as well so there is there is quite a big
disparity between whether they carry that gene and whether they're resistant or
not it's just a case of yeah there's certain certain tests you can do to check
that the genes get put on it's like mecca they like being grown 36 not 37 with a
little bit of salt in there as well and then then they become more resistant so
there's certain things you can do to test it a bit more. There are some reports
of AMR genes present in some kind of microbiome sequencing or some other
organisms just whole genome sequencing but when we phenotypically check it it
doesn't show any signs of resistance to that antibiotic so that's very
interesting because those genes are there but they're not active maybe their
promoter is missing or there is some silencing small RNA is being produced and
that's not letting them express it so towards your other point I just reminded
one that like do they need specific conditions sometimes as you mentioned they
like it salt or something yeah I don't know why just one of those things I've
just read somewhere okay and certain antibiotics you've got to have the addition
of extra metallo elements so daptomycin is often given to staph infections
especially like skin infections and stuff but testing it in the lab if you were
to test daptomycin resistant salmonella compared to staphylococci you've got to
add calcium and certain extra things to make sure that the gene the proteins
that they they make are have got enough you know oomph and have enough things to
make the actual protein work how do you know what you are growing in the lab is
actually what's causing the person to be sick so I look at a lot of carriage so
they're not always associated with disease ones that are associated with the
disease are usually taken from blood cultures so they're usually quite unwell
your blood should be fairly sterile so microbiome is very different I think it's
hard to define like if someone's sick in their microbiome and it's to do with
their microbiome like which one's causing it which one's not but from blood
cultures and usual sterile sites like CSFs and cerebral spinal fluid they should
be pretty pretty sterile so if you've got stuff going out there that person's
not having a great time lovely so do you think that this conditionally
expression of the genes it could be related to their pathogenesis or could be
related to their particular survival or is related to AMR yeah it could well be
so some some staff of genes to make them what we think let them survive in the
gut a bit more but they have to have high amounts of metal so they have metal
metal scavenging proteins like siderophores and zincophores and stuff like that
so that makes sense that you know some proteins that confer resistance need
extra metals and stuff so they will have those genes to help make the proteins
work thank you yeah do you think that the core genomes that we identify does it
has to be same in the gene whole genus or that they can be different in genus or
it can go more than a genus the core genome is same in all the bacteria if you
take any random bacteria like gram positive gram negative and you throw them
into a pan genome you'll get probably 150 genes in common because bacteria are
defined by something that a common ancestor billions of years ago but they are
do have some kind of shared mechanisms that make them just function and you know
even if you take your coli and salmonella they're separate about 100 200 million
years and still they have about 2000 genes like core genes that are in common
like it's very very high and that's why you can get transfers between a lot of
the gram positive gram negatives because they are so similar but there is a you
know a whole host of stuff out there that we have no idea what it does or how it
operates or what it does like if you look inside you like the microbiome inside
you a lot of it is uncultured people haven't done experiments on it it's because
it doesn't cause disease so no one really cares about it and so if you look at
anyone in this room like you're going to have probably novel species in genus
inside you that no one else has um which is kind of cool well no one no one else
has studied you know and so maybe we might sequence your poo at some point and
then you get to name something after yourself you know there you go unethical
but you could probably do that but as for core genomes yeah they they can vary
widely and depends on what you put in um and nowadays people are have gone from
say doing MLST which is where you look at seven genes and type them to core
genome MLST which is where you um build schemes for a you know a whole species
and then to pan genomes where you just build a species as you need a free to
study or you build pan genome for each study that you need and apply the core do
you think the essential genes like they are the same as core genome or how do we
identify the essential genes basically that's a good question so i think depends
on what your question is to begin with like what is the essential genes for that
situation you're looking at i think it's more of the idea of essential genes so
like heather was saying like different environments may need essential different
gynhyrchiadau cyffredinol, ac efallai y byddant yn arwain at bethau i'r
amgylchedd hwnnw, yn cymharu â'r rhai eraill. Mae'n rhan o'ch hanes
genedlaethol, yn ystod yr hyn sy'n angen ar gael yn y diwrnod heddiw. Mae'n rhan
o'r rhan hwnnw, rydw i'n ei ddweud. Iawn. Felly, Heather, rydych chi wedi'i
gynhyrchu hunedain o genomau Staphylococcus? Mwyaf, ie, technically. Iawn, mwyaf
o genomau Staph. Felly, a ydych chi wedi cydweithio â'r genom cyffredinol ar
gyfer Staphylococcus? A ydych chi wedi edrych ar yr angl hwn? Felly,
Staphylococci yn gynhyrchu. Mae pawb yn meddwl am Staph aureus, MRSA, ac rydw i
ddim yn gobeithio am hynny. Mae pawb yn ddiddorol am hynny. Felly, rydych chi'n
cael y ddau grwpiau. Rydych chi'n cael eich MRSA neu eich Staph aureus, ac yna
yr hyn rydw i'n ei ddweud o'ch grwpau non-aureus. Rydw i'n ei ddweud o'ch grwpau
non-aureus, ac yna yr hyn rydw i'n ei ddweud o'ch grwpau non-aureus. Ac maen
nhw'n cael eu cysylltu â'n gilydd. Pan ddywedais i'n ddiweddar, un wythnos yn
ôl, mae 83 o ddifrifolwyr mewn un grwp sy'n cael eu cysylltu â'n gilydd. Maen
nhw'n cyffredinol. Felly, os ydym ni'n dweud... Felly, os ydym ni'n dweud...
Felly, os ydym ni'n dweud... Os ydym ni'n dweud hynny yw un dyn, ac rydym ni'n
cymharu hwnnw â'n gilydd, mae'n rhywbeth o 75% i 78% o dyn. Felly, mae hynny,
fel y mae'n ymwneud â'r dyn, yn ymwneud â'r gilydd. Felly, maen nhw'n
cyffredinol. Felly, beth yw'r genome gwahanol rhwng un dyn a'r gilydd? Byddai'n
yr un peth fel edrych ar beth yw'r genome gwahanol rhwng un haemolyticus a'r
gilydd. Mae'n ddifrifol iawn ac yn gwahanol i gyd. Ac o'r drifoedd, byddwch
chi'n gadael un genome gŵr. Ond ar y cyfan yr holl ffyrdd o gynigion, mae...
Mae'n ddifrifol iawn. Ie, dydych chi ddim yn gallu darllen un. Ond yna, a fyddai
Emar yn dweud, dydych chi ddim yn edrych ar ddau? Mae'n Emar a'r 15 oed, ac yw'r
USA 300. Dyma oherwydd, yn ysbytai, dyw pobl ddim yn gwybod am rywbeth eraill.
Felly, mae'r rhwythennau hynny'n ddifrifol? Mae gennyn nhw ddifrifol am ail. Ac
mae rhai eraill sy'n cynnwys rhwythennau a ddifrifol, ond oherwydd eu bod yn bob
le, yn aml iawn, maen nhw'n dweud, oh, efallai, roedd e'n dod i mewn o, chi'n
gwybod, o ddod i mewn i ddod â chwaraeon llwyr, efallai, roedd e'n dod i mewn i
ddod â'r dŵr. Ac yna, felly, mae llawer o ddifrifolaethau ac ychydig o broblemau
o ddefnyddio ddifrifol, ac nid gyda Staphylococci, oherwydd maen nhw'n bob le.
Felly, mae'n aml iawn i chi ddod i mewn i ddifrifolaethau neu, fel, mae pobl yn
dweud, wel, nid oedd y blaid hwnnw wedi cael rhwydweithiau cyffredinol, felly
nid yw'n gallu cael ddefnyddio ddifrifolaeth. Ac rydych chi'n dweud, wel,
gallwch. Nid yw'n bob amser ddifrifolaeth pwntio sy'n rhaid ei greu a'r peth,
felly, mae'n amlwg. Fel y dywedoddwch chi, mae gennym ddifrifolaeth gwahanol o
ddifrifolaethau. Ac mae hynny'n mynd i fynd i'r cwestiynau i fy meddwl, a yw'n
bosibl bod y ddifrifolaethau genedlaethol sydd wedi'u cau'r ddifrifolaeth yn y
genome gynllunol, fel, sut rydym ni'n ei ddefnyddio ar y cyntaf, mae'r
ddifrifolaeth hwnnw yn ddifrifolaeth genedlaethol a'r ddifrifolaeth hwn yw'r
ddifrifolaeth genedlaethol neu'r ddifrifolaeth o'r cyntaf neu rhywbeth. Rydych
chi'n edrych ar y gwahaniaethau, felly, mae'n, mae'n rhaid i chi gael grŵp
cyffredinol, lle rydych chi'n gwybod efallai yw'r ddifrifolaeth, ac yna rhywbeth
sydd gyda ddifrifolaeth, ac dyna lle rhaid i chi fynd a ddifrifolaeth ac mae
angen data ddifrifolaethol i gefnogi'r holl hynny. Dyna lle mae'r lab yn
gweithio gyda phobl yn ymwneud â gwneud analysiadau mathematig, oherwydd os nad
ydyn ni'n deall beth sy'n digwydd yn y lab, yna byddwn ni'n creu gwirionedd. A
gallwch chi bob amser ddod o'r cyfnod, ac mae angen i'r holl ddifrifolaeth ddod
o'r cyfnod ac dweud beth sydd yn digwydd. Ond mae'n dweud, dyna'r un peth. Mae'n
dweud, mae'n dweud, mae'n dweud, mae'n dweud, mae'n dweud, mae'n dweud, mae'n
dweud, mae'n dweud, mae'n dweud, ond bawb yn dweud. Ond dydyn ni wedi dod o'r
Cyfnod yn ddiogel mae rhwng hystod dau fel cyfnod tŷ mathristio a rhagor i
bwysio'ríoedd sy'n dod at hynyn. Yn gwirionedd, mae'r rhain yn hollol queithau?
Ac a allu iddyn eu defnyddio cyflawn am bod y mathau genedlaethol hyd yn oed
fiw? Ie, yna sydd llawer, fel maen nhw'n dod o'r cyfnod genedlaethol, ac mae
llawer o'r cyflawn sy'n cael eu gweld yn gwirionedd yn gwneud dim byd. Maen
nhw'n eu gofyn cyflawn cryptig. Nid ydyn nhw'n gwybod pam maen nhw'n yno neu pa
ffordd gyffredinol mae'n rhoi i'r bacteria, ond mae cost i'r holl hyn, ac felly
mae'r bacteria yn mynd i'w ddod allan. Maen nhw'n mynd i fod yn cael eu rhannu
drwy ddatblygu yn ystod y cyfle. Ond yn aml, dydyn ni ddim gwybod beth maen
nhw'n ei wneud, nid ydyn ni'n gwybod, er enghraifft, yn y byd, beth sy'n ei
wneud pob gynid yn hynny. Felly, fel y K12, rydyn ni'n debyg pan mae'n mynd i'r
holl bethau eraill. Ond mae llawer fawr o wybodaeth genedlaethol yn cael eu
rhannu bob amser mewn microbiome gwahanol. Dydyn ni ddim ei weld oherwydd mae'n
cyflawn sy'n gwneud dim byd ac nid ydyn nhw'n eu astudio. A ydych chi'n
defnyddio transbosonau neu elementau genedlaethol mewn unrhyw ffordd o
weithgareddau neu astudiaethau gwahanol? Rydw i'n hoffi siarad am transbosonau.
Ond dydych chi, ydych chi. Iawn, dyna'r cwestiynau i mi. Iawn, beth yw'r
transboson a phwy ydych chi'n ei ddefnyddio? Mae transbosonau yna, efallai eich
bod chi'n clywed, efallai eich bod chi'n clywed yn gyntaf biologi GCSE gyda'r
corn, yw hi, y corn bleu a hawdd. Ond cael owirion o DNA. Gallwn gwneud eu
humaniaeth i fynd ar yr HAB er mwyn gwneud rhomaidd o dein gene i gwneud
offline-ymgyrchu fel diwydiant o ddefnyddiant yn ddiwylliannol o'r enw Tradis
pan edrychon ni ar gysylltiadau transposon Fy hoff iawn Fy hoff iawn Rydyn ni'n
ceisio gysylltio'r transposon yn yr un ffordd o gwirionedd i sefydliad o
bakteriaeth Gweithio nhw i mewn sefyllfa ac yna os yw'r geneau'n bwysig mae'n
rhaid iddyn nhw fod yno dydyn nhw ddim yn gallu bod yno gallwch chi yna gwneud
eich rhamaidd genedlaethol ar gyfer pob un ohonyn nhw ac yna edrych ar y geneau
sydd yno maen nhw'r pethau pwysig y pethau sydd wedi cael eu rhannu gan y
transposon nid ydyn nhw'n gallu mynd felly maen nhw'r pethau rydyn ni'n ei angen
y pethau pwysig OK Felly mae'n golygu y gallwn ddefnyddio'r elementau
genedlaethol bywyd ar gyfer ein bod yn dda hefyd ar gyfer biotechnolegiaeth yn
deall'r mecanismau nid ydyn nhw'n iawn felly mae yna rhai dda hefyd neu gallwn
ddefnyddio nhw ar gyfer y ddau dda Iawn Felly, a yw'n bosib i ddefnyddio trwy
gwneud seicwensiau genedlaethol pethau rydyn ni'n eisiau gwybod neu trwy gwybod
y ffengsiynau mae angen i ni wneud unrhyw beth eraill o'r elementau genedlaethol
bywyd unrhyw un Felly, rydyn ni'n gwneud seicwensiau a rydyn ni'n siarad am
seicwensi o'r ddau dda hefyd ac rydyn ni'n gwybod y ffordd o wneud y seicwensi
oherwydd rydyn ni'n gwybod y strwythur ond os ydyn ni eisiau gwybod y
ffengsiynau genedlaethol beth rydyn ni'n rhaid ei wneud sut y gallwn ei
ddarganfod Ysgrifennu yn y lab Cysylltu'n allan gweld, ie os ydych chi'n cael
syniad gallwch chi edrych ar os ydych chi'n cael y seicwensi gallwch chi edrych
ar syniad genedlaethol syniad genedlaethol a oes ganddo homolog yn unrhyw beth
eraill a oes hynny'n edrych ar rhywbeth eraill y gallwch chi'n testio arnynt ond
mae rhai genedlaethol a phroteinau sy'n dweud nid ydyn ni'n gwybod mewn
gwirionedd Wel, gallwch chi nawr ddweud nad ydych chi'n gwybod ac mae llawer
wedi'u hyfforddi fel proteinau hypothetig Dyma fy bywyd o edrych ar genedlaethol
staff mae'n hypothetig Ie, dwi ddim yn gwybod ffengsiynau pob genedlaethol wrth
gwrs fel ein bod ni'n gwybod unig y fracsiwn o'r genedlaethol, mewn gwirionedd
rydym ni'n gwybod er enghraifft mewn ecoli rwy'n credu rydym ni'n gwybod y
ffengsiynau o hanner o'r proteinau sy'n cael eu hyfforddi Ond mae'n testio
pethau gyda'n neges positif felly rwy'n edrych ar lawer o staff lecochlai yn
cario felly nid ydyn nhw'n gyflawni dymor maen nhw wedi dod o'r hyn ond yn
cymharu y rhain sy'n cario y rhai rydyn ni mewn gwirionedd ac yn cymharu iddyn
nhw gyda'r rhai sy'n cael dymor mae'n edrych ar y genedlaethau sy'n gwahanol
rhwng y setiau hynny sy'n gallu helpu edrych ar pethogenesiaeth a phactorae ac
arian Mae'n edrych ar y gwahanol a'r cyfathrebu mewn grŵp da a grŵp ddau sy'n
helpu i ddefnyddio neu'n helpu i chyfathrebu ar y rhai sy'n cael broblemau Rydyn
ni wedi siarad am beth yw'r genedlaethau cyllid a pheth yw'r genedlaethau cyllid
a sut maen nhw'n gallu cyfrannu at mecanismau'r cyfathrebu neu'r cyfathrebu'r
cyfathrebu Felly yw'r phage yn gallu cael eu defnyddio ar gyfer y cyfraniad
hefyd fel rydych chi'n siarad am y transposonau mae'r phage yn ffordd o
genedlaethau cyllid a allu cael eu defnyddio fel unrhyw fath o cyfraniad
cyfraniadol neu rhywbeth Mae llawer o edrych ar phage a ddarparu adnoddau a
pethau ar hyn o bryd Nid ydw i'n berson phage Ond mae... Rwy wedi gwneud ychydig
o waith phage yn fy bywyd ffug Ond ie Roedd yn fwy... defnyddio  phages to
tackle bacteria, so ones that actually attack bacteria, so then hopefully you
could clear an infection with phage treatment instead. Okay, so phages are not
only involved in the spread of antibiotic resistance, but they also can be used
to control it. Okay. Well, one problem with phage is obviously the bacteria have
a mechanism to record what phage is trying to kill them and embed it in their
genome and then that's it. It's resistant to that phage. The CRISPR system.
CRISPRs, yes. We all know about that. But not all bacteria have the CRISPR
system, but yeah, that is another implication. There's always one in there that
will cause a problem. So in the past, if we look just like two decades before,
like the situation was quite different than it is today. We didn't have the
facility of sequencing the whole genome of the bacteria, and we were totally
relying on the phenotypic observations. With this advancement in the technology
and the sequencing amount of data we got, is there a way forward, do you think,
that we can use this amount of data and the technology to tackle these
antibiotic resistance mechanisms? Not really. We need to stop using antibiotics
for inappropriate purposes. All it takes is one person in one part of the world
to inappropriately use antibiotics and then it can have devastating consequences
around the world because a lot of these introductions are caused by one instance
and one crossover of, say, extensive drug resistance, and then it spreads. So
it's a global effort that we have to fight. Okay, no, that's a very important
point because we all should be vigilant and we should all be responsibly using
the antibiotics. But my question was, is AI or machine learning can help us to
tackle these resistance mechanisms or identify resistance mechanisms and stop it
before they happen? I mean, my PhD was in machine learning, so I understand AI
quite a lot, and I know that it's a buzzword at the moment. Everyone thinks it's
the solution to everything, but it's not necessarily. It's good for certain
applications, but not good for other applications. And as long as you have high-
quality data, you can start mining it. But at the moment, we don't really have
that kind of data. I could envisage some very interesting PhD and postdoc
projects for doing this kind of stuff, but I don't think it's going to be as
useful as people claim it's going to be useful. They might get funding from a
silly VC who doesn't know any better, but they might be bamboozled by all the
buzzwords. But I'd be cautious. Even with that, you've still got to, again, test
it in the lab, test it in the situation. So it's always going to be a two-way,
like, you can't just let a bioinformatician tell us the answer, because it will
still need testing in the lab, and likewise, we're going to do stuff in the lab.
We're going to need a bioinformatician to look into it, so it's still going to
have a two-way thing. It's not going to solve the world. We still have to test
it. Okay, so we all agree that we should be using antibiotics responsibly.
Actually, I think the next big step is actually for sequencing is going to be
culture-free diagnostics. So when you go into a hospital, then rather than
spending a day or six weeks culturing a bug to find the antibiotic resistance
profile, it can be done quickly with, say, Nanopore. You get a result in a few
minutes. It'll tell you what antibiotics may work, what antibiotics may not
work, and then they can appropriately prescribe drugs rather than doing this
kind of trial and error that they do at the moment. So I want to do a little bit
of kind of survey in the audience. Have you ever taken antibiotic? All of us
have, okay, at some point in our life. Have you visited physician and you
thought you need antibiotic, but physician didn't recommend to you? You thought,
like, you feel like, okay, I'm feeling very miserable, but I need antibiotic. I
have a throat infection or whatever, but physician didn't recommend to you.
Okay. I did feel like sometimes I go and talk to physician and I had ear
infection once and the physician didn't prescribe me the antibiotic. He gave me
some topical spray that was not antibiotic. It was just... It's acetic acid
usually. Yeah, exactly. It's kind of... Similar to ear and eye infection. So,
you know, people are using it responsibly, and I really appreciate the audience
here. They are really responsible. So next, we think about, like Salmonella,
that we have developed XDR typhi, it's the cause of infection in developing
countries, but we see the increasing number of cases in UK and USA, in the
developed countries, Europe as well. So what do you think, how can we stop that?
How can we minimize that or contain it? So this is where I think sequencing is
great at the minute because you can monitor a situation. So since the outbreak
was originally identified, we can see it traveled to many different countries
and you can monitor that really well. It may not solve the problem, but I think
it makes people aware of the problem. So it makes the people that need to get
things done aware of what's actually happening. And that's been started to put
in place with vaccine programs actually in those regions. So because they know
about it now, they can actually implement a vaccine which hopefully will help to
eradicate it in the future. Although the solution is just a proper sanitation
system. That would also be very useful. Of course. As you mentioned a very
important point, long-read sequencing or MinION. So do you think use of MinION
or long-read sequencing can help us in identifying transposons or mobile genetic
elements more effectively? It can and it can't. Initially when they started
doing long-read sequencing, they found actually a lot of the mobile genetic
elements were disappearing. And it was because the original algorithms, which
some people still use, would take the short long-reads, short fragments, and use
those to correct the longer fragments. And then algorithmically, it would say,
OK, you know, anything below 3000 bases will be used to correct all the other
long ones. So you get higher quality long bits. But unfortunately, some mobile
genetic elements are very short. And so you'd have disappearing plasmids. And
that's initially what they found was happening. So you just got to be aware that
bioinformatics doesn't solve everything and it can introduce extra errors and it
can actually make you miss things. And you might be blindly looking at your data
and not realizing that it's just the algorithms are making it disappear. But
actually, long-read sequencing will be very useful. And where it's most useful,
I think, moving forward is with, if you can take an uncultured sample. So I've
said this already, but you can do like adaptive long-read sequencing. Are you
doing that? No, but it is very rapid. So there are methods that can, instead of
doing the culture method, which would take two days or longer, it would actually
take potentially six hours from the person to actually get a result to say what
it is and what is in it. So you can treat it with the right antibiotic at that
moment in time. So like with TB now, rather than spending six weeks growing up
TB, TB is terribly antibiotic resistant. Like some strains are totally drug
resistant, isn't that it? And so that's a big problem, obviously, because TB is
very widespread and it spreads very, very easily, unfortunately, and the vaccine
is pretty poor. And so now they've switched over to doing sequencing. So instead
of spending six weeks taking a result in a week or two, then appropriately
provide treatment, which the treatment can take, you know, six months or 18
months, depending on what strain a person has. So that's quite a good thing, you
know, that's there to tackle antibiotic resistance, people being treated
appropriately, not being treated blindly. Yeah, and I totally agree with you
tuberculosis is a big, big problem because of their longer growth, doubling
time. So they take ages to grow as compared to E. coli or salmonella or
staphylococcus. And that's why when we try to test antimicrobial susceptibility
on TB, it takes six weeks or something like that. Yeah, the normal growth time
for that is about three weeks. So if you, I think if you don't grow in the right
conditions, and it doesn't grow on that certain time, then you've lost all that
time already and you have to start growing it again. And then you could miss it
again. So that could be a continuous cycle, which could be solved really easily
with sequencing. And hopefully over time the price will go down so it can be
implicated more in developing countries where you get a lot of multidrug
resistant tuberculosis and stuff like that. Because I know where I've worked in
Uganda for a short period of time, and there was a TB clinic there and there
was, I had the one computer of the village. So when they've got a lot of
multidrug resistant tuberculosis there, there was no means to do testing. Yeah,
there's a lot of implications in developing countries sometimes, especially with
antibiotics not being what they are and I think WHO guidelines at that time.
50% of antibiotics were actually what they said they were. So it is a global,
when we're thinking about AMR and resistance, it is a global thing. We need
everyone to be on board, everyone treating everything, which is hard. The
pandemic has been very good actually for low middle income countries because now
sequencers have been spread around the world. Like we sent out minions and a
laptop and everything to Zimbabwe, to National Reference Lab there. And so they
then did their first long-range sequencing because of COVID. But that means they
have all the training and equipment there to then do it for salmonella
typhimurium, for example, and typhi, which is what the original project was for.
So, you know, you've got capacity that's been redirected appropriately to
bacteria, which is obviously better. So mobile genetic elements, they are bad,
but they are not the only culprit for all the antibiotic resistance mechanisms.
Am I right, Andrew, about that or not? Yep. And if you use them, Yasir, you're
trying to use them for good. Oh, we can use them for our good. Not only the
transposons, but other plasmids as well, because we use plasmids and other
mobile genetic elements to make certain kind of molecules that are used for
good, like insulin production. We kind of produce insulin and other proteins
that are being used, and we use them in our clinical settings, in pharmaceutical
or even industry, like in the households, like there are some enzymes that are
produced as detergents. We use them and they are all used one way or another. We
employ mobile genetic element to produce those kinds of things. Thank you very
much for the discussion. It was very beneficial. And as I mentioned earlier,
that antibiotic resistance, it does spread with the help of mobile genetic
elements, and it spread through vertical means and as well as horizontal
transfer of the genome across different species and different general. So mobile
genetic elements are the part of a major problem, and our sequencing that we
use, luminal sequencing or the longer sequencing that we use, they are very
beneficial because they can educate us about the situation very quickly that we
were not able to do that a couple of decades ago. And because of that, we have
been able to contain and not only the antibiotic resistance mechanism, but we
have also been able to manage the disease by producing certain kind of vaccines,
targeted vaccines that are with the help of, of course, knowing the sequencing
and the structure of the proteins and genome sequencing and the mutation that
are happening in the real time. So I must say that the advances in technology
that have brought us this far, they can be used for the beneficial fishing,
beneficial fishing of the humanity and the health system in future. Thank you
very much for your contribution. It was lovely talking to you. Thank you. Thank
you very much. Thank you. Thank you so much for listening to us at home. If you
like this podcast, please subscribe and rate us on iTunes, Spotify, SoundCloud,
or the platform of your choice. Follow us on Twitter at Microbinfee. And if you
don't like this podcast, please don't do anything. This podcast was recorded by
the Microbial Bioinformatics Group. The opinions expressed here are our own and
do not necessarily reflect the views of CDC or the Quadram Institute.