Applications of Next-Generation Sequencing (NGS)


So welcome back to this course functional
genomics So in this lecture we are going to look into some of the applications of next
generation sequencing So in the previous lecture we have seen the principle behind the sequencing
and the few examples like cancer genomics So we are going to go beyond cancer and look
at what are the other applications one could think of using this powerful tool of next
generation sequencing So here is a paper it is again a landmark
paper which sought of established the power of next generation sequencing Here they have
developed or they have shown the sensitivity at which the sequencing can help as a screen
for prenatal measurement of fetal genome So you must have read about disorders that affect
the you know children For example mental retardation is one such disorder there could be many other
disorder like red syndrome and others it happens because de novo mutation meaning the gambit
that led to the embryo had a mutation as a result the baby is having a mutation This is unexpected because the parents are
normal but the baby could be and likewise in down syndrome and as the age of the mother
at pregnant mother increases the chances of that developing embryo might have down syndrome
increases So the traditional method is to look at the amniotic fluid from the uterus
and go for screening either it is a chromosome or DNA sequence analysis and so on But that
also puts the baby at risk because these procedure involved requires invasive procedure So if
you can come up with certain noninvasive screening methods to understand the genome of the growing
embryo that is going to really help And with the cost of such sequencing coming
down it can be a routine screening for you know the developing embryo to see whether
the embryo has any genetic defect before you know certain development stage where you know
the parents could take a decision as to whether they would like to have the baby or not which
is legally allowed So this paper talks about either noninvasive prenatal measurement of
fetal genome what is that So what is known is that for long it has been
suspected that you know some of the DNA or RNA of the fetal fetus the growing embryo
can be found in the circulating blood of the mother So it can come even few cells can come
because the umbilical cord where the exchange of blood takes place there could be few cells
coming into or it could be only the DNA coming only the RNA coming because there are process
by which you know there is a now we know that RNA output into small vesicle they are transported
and so on So what they have done is that they have used
a method that is the tube maternal peripheral blood And then captured the DNA that are present
in the free DNA that are present in the plasma liquid part of that blood and they have done
exom sequencing ok And likewise they have done a shotgun sequencing for the plasma free
DNA and then looked at the sequence So what does it help so this the idea is that
if you have the blood carry some RNA or DNA from the fetus the growing embryo than by
doing this sequencing where you are sequencing every fragment that is captured and this is
so sensitive because these are clonal meaning that each cluster represent one copy of the
DNA we are able to sequence each cluster get the sequence separately You may find that
you know if you know that the mother carries the genome from his father her father and
her mother So you know that is one and ofcourse is a
complementary what is shown here but you can really type that what is shown here So this
is the mothers DNA which is representing her for example father this is the mothers DNA
that represent the mother and these are the sequences which are different between father
and mother these are polymorphic side which you are able to genotype and call from fomites
come from and obviously the fetus would have DNA sequence that had come from her father
right So one that has come from mother the other
copy that had come from father so this is the information that you have you have the
DNA sequence of the father you have the DNA sequence of the mother which tells you what
are the that they have which has come from his father and his mother and so on And then
you have fragments of the DNA that are sequenced using this approach and all you have to look
for is that you know you are there any genome that are representing the signature of you
know father which clearly tells that yes that there are genome that represent the father
that certainly should have been form the fetus because is coming out So that it tells that you know there are circulating
you know free DNA derived from the fetus so that is the approach that they have used and
they are able to really show that you can do genotyping of the fetus by looking at the
free DNA that is there in the serum right and this is powerful tool because you can
type you can look at the mutations and you can look at the copy number and say whether
it is a down syndrome and so on So that is really going to change even with some more
improvement of the technique it can become a routine screen in the future it is a possibility
such is a huge advancement The other application that can you know that
is of great interest is the thousand genome project here the idea is to sequence complete
genome of 1000 individuals and to look for variations something that we have discussed
about And these are individuals that are not restricted to particular race they have selected
from different continents like what is shown here and they are trying to compare and look
at what variations exist amongst the population The idea is to create what is called the reference
sequence say suppose a mice genome sequence has been sequenced Now when I when this sequence has been done
I need to compare the sequence with some reference sequences to infer what is the variation whether
the variation is something unique to me or present in the population So for this to happen
you should have enough number of samples already done and in present in the database So that
is the objective of the project So you have atleast hundred plus you know individuals
DNA sequenced and kept in a database representing each population so as in when such kind of
technologies available for individuals then they are sequenced and then you can immediately
match with the reference sequence and say whether the variant that was found is common
or not common whether which segment of the genomic represent whether it could be deleterious
and such kind of you know analysis can be done using this reference project So these are the images representing the next
generation sequencers you can see they become very small table top models have come in The
different companies that are you know marketing these products and they have the chemistry
and as well as the two you know the kits for preparing the DNA and ligating into adaptor
and so on So these are some of the popular model that you have This is just to show you how things have changed
so you know we started with our discussion on the manual sequencing how the extra film
was used to understand the sequencers to the high throughput capillary sequencers and now
what we are talking about the next generation sequencing and what we are looking at very
soon is a product called MinION this is all that you have that is very small device that
you can keep in your palm it is a personal sequencer So it is like just like your cellphone
the USB cord you can connect to the computer and little bit of your cell or your blood
or whatever and it would complete your genome sequencing in few hours that is what we are
expecting And just to talk about that we have a video and I will play that in the next slide
so where you can understand Oxford Nanopore Technoligies is developing
Nanopore sensing technologies with the analysis of biological molecules In oxford Nanopore
Systems the Nanopore is inserted into a membrane created from synthetic polymers A potential
is applied across the membrane resulting in a current flowing only few the aperture of
the Nanopore Single molecules that end to the Nanopore goes characteristic disruption
in the current this is a Nanopore signal by measuring that disruption the molecule can
be identified The MinION is a small portable device for
the analysis of single molecules such as DNA RNA and proteins It has been designed for
uses few want it simple fast plug and play device that can generate real time data not
confined to the laboratory Inside the MinION device is a flow cell this includes the censor
array which is a collection of electrodes and micro supports each one of which has an
individually addressable electronic channel Nanopores are built into membranes lying across
the micro supports ad it is here that the single molecules are analyzed using the change
in current passing through the Nanopore The MinION has a simple and fast workflow
and its portability and versatility of experimental design of its opportunities in a number of
different applications Firstly the MinION is connected to a laptop through the USB free
cable The use of is the instrument control software After performing a calibration and
quality control procedure the device is ready for use This sample is added to a port in
the MinION and fluid flows across the surface of the censor array and the user begins the
experiment The Nanopore signal is measured by an application specific integrated circuit
in the flow cell and processed by MinION the instrument control software This software carries out several code at
a tasks and can be used to change the experimental workflows or parameters The user receives
real time feedback of k matrix such is sample quality and number of events combined with
real time cloud based analysis from METRICHOR over the use of own systems analysis can start
immediately and continue the experiment until the point where sufficient data has been generated
to answer the biological question The system may be active for minutes or days rather than
being controlled by an arbitrary runtime and subsequent analysis this is known as run until systems produce full length full read whether
it is DNA RNA approach in real time in real time as the machine runs A running is in real
time as the number of advantages obvious one is which is compute cost and management overhead
but it also means you can look for things in real time If you can things in real time
you can then feed the back through system and you can move the whole system or part
of it onto another sample or other experiment Simple to use with minimum sample probe the
MinION is the world’s first portable device for the analysis of biological molecules For
more details please visit Nanoporetechcom How this device can work you have to understand
all these are commercial things they are not going to disclose everything that led to the
development of this device by but it would give an overview as to what is the principle
behind that And one can buy this is not going to be expensive and if it becomes really because
this has come like a cell phone and you can expect that this would flood if it is successful
it will flood the market and sequencing would become just like the blood pressure or just
like people measure the glucose level in your blood is going to be that easy so that is
the future right So having seen the video now let us look into
some of the advantages and disadvantage of the next generation sequencing so has is the
case with any advancement any technique or any device there are something good there
are something not so good The advantages of course very high throughput you know the number
of sequence reaction it can do is anonymous just put the entire genome it is able to sequence
and you can scale up you can multiply whatever very cheap the data production but it has
its own disadvantage The disadvantage is relatively short reads
so 50 bases 100 bases that is all it can read But what do you do is for the same segment
there are multiple fragments overlapping fragments you are reading Therefor you have to have
more reads for a given segment to you know match assemble and get the sequence And because
of this process is also relatively higher error reads because you know there are lot
of amplification process So every cluster we spoke about that we spoke
about that video we are shown that one DNA molecule stuck to the adaptor then you know
you have PCR like amplification The DNA polymerase goes over to make multiple copies and then
that copies are sequenced So when you are using this kind of polymerases they are not
you know unlike our cells where the polymerase copies and there is a mechanism that looks
at whether the copying process is accurate is there any mismatch So that does not happen here therefore there
could be errors it is pretty high as compared to the traditional Sanger method The third
is that the you know the data that it generates it is so huge that it really requires a very
good bioinformatic tool to assemble the sequences to analyze is much more challenging See in
traditional Sanger method anybody should be able to analyze the DNA sequence but for the
NGS platform the sequence that it generates requires really skilled person analyzing the
data So that is one of the cartoon that talks about
NGS machine that it is like dumps you know it dumps you know huge data sometimes you
wonder what you do that unless you are able to mine and get the one that you want and
others what is the noise what is the signal so that is something it comes out So there are many vocabulary people use some
of them in the next generation sequencing So we will talk about some for example the
read they talk about how many reads This is you know small a segment of the DNA how many
times that has been sequenced in independent unit independent clusters as a continuous
sequence produce by the sequencer right Coverage this is what we spoke about you know how many
short reads of overlapping segments you know that that are present Consensus sequence so when you have several
sequences you know sequences you assemble to get you know consensus sequence And this
is a Contig Contig is nothing but you have assembled a small segment little larger than
the reads but but they are not connecting to something else but represent much larger
as compare to the reads And then you know you have to align with the known sequence
to you know get the region for example whether it is DNA coding region and all other things
So these are some of the issues that that come So we are also talking about single versus
paired end sequencing directional unidirectional libraries I will discuss it little later So this is what it is so you have the reads
you know these are reads that represent small segments multiple of them you got the sequence
and then each sequence is aligned with overlapping sequence to form what is called as Consensus
sequence that represent the Contig But still the Contig you know there are gaps within
that but you are able to say that this Contig is located here in the the same you know physical
orientation that you are able to do by aligning the sequence or comparing it with the genome
sequence that is available So this is how you are able to map right So this is the workflow the data that is generated
you know how do you really study that so the analyses of the the database leaves image
analysis because every time base is added in the cluster it is going to emit a particular
light So but in a chip we are going to have hundreds of thousands spots for every read
So you know there is a powerful image analysis tools that cause for every cluster as to what
base has been update The way you are able to get put together this
sequence there are two different methods one is called as De novo genome that is you are
sequencing a genome and the sequence is not known previously ok So the way you analyze
the sequence is very different as comparing to a genome which is already being sequenced
but we are looking at what difference you have it in this particular genome For example
human genome is available I want to sequence my genome and identify what variants I got
So I am going to compare with some existing reference data so the way you analyze is different
De novo meaning you are generating for the first time So you have of course the DNA fragment and
then you have sequenced it and you are finding this overlapping segments and overlapping
segments you know to make Contig and the Contigs are looked at for the sequences and then you
make you know much larger like what we discussed Obviously this method is not going to give
you the continuous DNA sequence there will be gaps and one has to you know fill the gaps
with library and other other methods that is for sure But in the second method where
you are talking about analyzing the sequence for example you want to look into the RNA
sequence or you want to find where your transcription factor binds or polymorphism or mutations And basically you are reading you know single-end
reads one end of the DNA you are sequencing and then whatever sequence that come you are
comparing with the reference sequence and that is where you are compiling as to whether
sequence are identical or there any difference The other method is you have a longer DNA
fragments as I told you the reads are you know the length that you can sequence is restricted
here 50 bases 100 bases 150 bases whatever So you are not going to sequence the entire
DNA fragment but we are sequencing either end of the fragments and we are going to you
know align in like this that gives you this called as paired-end reads but in this we
are going to have segments that are not sequenced right So that is a limitation but this is good enough
for you to tell if you are sequencing this is an RNA cDNA rather it only tells how many
copies of the RNA is present So still it gives you information likewise it you are talking
about ChIP sequencing data what are the DNA elements on which the transcription factor
bound So how many times there is bound so we are talking about quantitative measurements
these parents reads are good enough because it tells you how many counts that are present So that is about the next generation sequence
and sequence analysis to some extent Now we are going to talk about much larger application
beyond you know the mutation detection Of course one is whole genome sequence resequencing
like what we have seen 1000 genome project and so on We also looked at for example transcriptome
analysis and ofcourse we can try to understand how the genes are regulated in terms of changes
in the DNA in terms of transcription factor binding to that and and and epigenetic changes
for example methylation of the DNA or metagenomics we discussed where we are looking into large
number of microbes we are sequencing classifying them based on certain signatures or paleogenomics
for example pieces you know you can sequence it quickly and understand the sequence and
and so on So these are examples one I am going to show
you is what is called as RNAseq popularly known as for sequencing RNA right So let us
say you are looking at two conditions condition one is a tumor condition two is a normal skin
Let us say the skin with the tumor and skin that is not having a tumor You isolate RNA
exactly the same way like traditional method and then you are going to pull out all the
MRNA if you are looking into the coding RNAs and then obviously the first step would be
convert the RNA into a DNA what you call as cDNA fragment and make it double stranded And then just like the way you have done it
for DNA you add you know adaptors on either end of the DNA this is what you do it right
And then the application is exactly same like DNA sequencing you put through the small channels
and then allow the DNA to go on bind and make clusters and then use one of the chemistry
to you know read the sequence that is what shown here as a dot like the way it is explained
in the YouTube And then you are going to analyze the sequence
so it could be you know short inserts or you have reads on either end of the DNA and once
you get the sequences you are going to go and you know align with the the genomic data
that is shown here for example that would tell you a given segment of you know transcript
has come from representing exon and intron and and there is spliced product or unspliced
product all this information can be obtained So this is how you do that further so you
have the raw data and then you have aligned them to make the Contigs or assembly And then
you you know compare them with the reference sequence that talks about for example Junction-crossing
read count for example alternate splicing right Exon one is joined with exon three or
exon two or exon four So by looking at the sequence in the exon boundaries you will be
able to tell how many times how many different ways the exon is spliced or you can talk about
fusion events Sometimes your chromosomes are trans located
as a result a two different gene is fused together therefore you have a what is called
as a transcripts have been different coding regions easily you can look into that or you
can look at even what is called as allele-specific expression meaning you have you know for all
the genes that are present in your autosomes you have two copies one representing father
one representing mother For some of the genes it could be only the
maternal gene copies expressed meaning the gene copy that you derived from mother is
expressed not the gene copy that got from father So how would you really do this so
you can sequence it and there are variants that are present where these variants allele
could be different between father and mother By looking at the sequence of the RNA we will
be able to tell is there that both the paternal and maternal allele contribute or equal number
of transcripts or there is a disparity you have more of from the maternal one or more
transcripts from the paternal If there is a difference then you can go back
and look into the genome organization and sees there anything that affects the way the
transcription is being done So that is another powerful tool which is normally you miss out
when you do a micro array or real time PCR that we discussed earlier And then we can
talk about other you know we can talk about count matrix meaning how many reads you have
got for you have gotten for a given run transcript of genes that talks about the quantity the
expression difference So if you have a normal skin versus tumor skin then normally we talk
about different ways so on But you can also talk about the differential expression and
and then you will be able to calculate come up with that And then since you are looking at a large
number of RNAs at the same time you can model it if gene A expression goes up what happens
to other genes is it that there are some genes whose expression also goes up so you have
large number of samples we can do model and then we can we will be able to tell whether
such kind of you know co expression patters you can identify which help you to predict
more networks and so on right This is just to give you some example as to
how people do this is RNASeq data so these are nothing these mountains represent the
reads how many reads happen for you know the region of the DNA you can five payment of
the gene three payment of the gene exon 1 2 3 4 5 6 and this is UTR You find that the
number of you know reads that you found for each segment So with this way we are able
to map it not only that we can go on and talk about for example if we go for other methods
for example sit in a library or other methods normally your analysis power more restricted
to the three payment which sought of you know is taken care in case of the RNASeq So it
is one of the major advantages I gave you example of chimeric gene when there
is a translocation it could so happen that to different genes fused together by looking
at the transcript Now here you can find out such kind of transcript where this pricing
you know took place such that you know two different genes now transcript and joint together
So this is a powerful method that you have The last section that we are going to discuss
is how the chromatin or the DNA modifications can alter the way gene functions right What
are the way the transcription can be regulated the cartoon represent here the DNA and then
you have what is called as enhancer sequencers onto which there are some proteins comes and
bind is called as enhancer activating proteins And you have of course the transcription factors
which bind to and then is all together is a complex to RNA polymer is for the gene to
be transcribed So now whether a gene is expressed or not
expressed depends on whether these transcription factors are present And whether this region
of the chromosome is open for the protein to come and bind It happens because of many
factors one of them could be the changes in the DNA which is called as methylation or
at the chromatin it could be histone acetylation and so on And how do you really look at such
kind of changes at the global level So one way people do is to do what is called
as chromatin immunoprecipitation So basically what you have is that you have the chromatin
here histone shown that has tail coming out which can be modified It could be acetylation
it could be methylation it could be ubiquitination it could be sumoylation it could be phosphorylation
So if you have an antibody which recognize acetylated histone or methylated histone or
any other modification specific at antibody would now come and bind to this region So now if you can use this as anchor to pool
the DNA that are present close to it then and then sequence the DNA it will tell you
which was the regions that are in the chromatin acetylated or methylated so this is the way
Or I could use an antibody against anyone of the transmission factor Now and if I pull
using this antibody the DNA that are bound to this transcription factor it will tell
me as to which segment of that genome the transcription has bound to So this is a powerful method and this is what
called as chromatin immunoprecipitation The schematic is shown here so what basically
it do is there you isolate the chromatin and then cross link the proteins to DNAs So can
you chemical moieties which stabilizers what is shown here so the DNA and protein is stabilized
and then once you have done that what you do is you saw of the DNA net you know cut
the DNA randomly therefore you have protein complex in which you have the DNA bound to
it so there are going to be Now if you use an antibody which recognize
a given protein for example I am looking at transcription factor A Now this is the antibody
that binds to transcription factor A so it is going to pull wherever transcription factor
is there in the protein it is going to pull if the transcription factor is bound to certain
DNA that DNA also come along with that after which I extract the DNA and I sequence it
and with the sequence I can go on and map the genome to see which region that sequence
represent therefore I can tell this is a region onto which the transcription factor is bound
to So this is the way people do And this is one such landmark paper where
they have looked at high resolution profiling of histone methylation of the human genome
So they have used this chromatin immunoprecipitation followed by deep sequencing to understand
how it happens So basically they have pulled the DNA using ChIP immunoprecipitation and
then they went and did this NGS you know you have small fragments that are derived from
the ChIP DNA you added the adaptors you pass through the fluids the device and then we
make clones we make this clusters several copies of it and then went through this sequencer
to get the data So with that now we are able to tell for example
the regions in the genome where the methylation is very high because you use an antibody now
recognize for example methylation of the you know histone Now it tells you for example
this segment of the genome you have severally you know reached that means in many cells
this is a segment onto which you know the antibody is able to bind which suggest this
is region the histone is methylated What is that region you know it is an upstream region
or downstream to upstream to a particular you know gene So it tells you that this region is methylated
and this is you know expressed or not expressed you can go and analyze later And this is going
to really help because we are talking about epigenetics Epigenetics is I may have a normal
DNA sequence but the way my genome functions is modified because the kind of environment
I live in for example right So if I can understand now what are the regions that are you know
of the chromatin that are methylated is going to help how the environment modify the genome
its function without changing the base So that is a huge huge advantage in understanding
the function And some of other you know sequence shown
here for example this is an active gene this is an inactive gene or a condition in which
the gene is inactive a condition in which the gene is active You can see that this is
this is the number of reads that are denoted by the line diagram here you can see the enhancer
region if you have used then you can find that the methylation is altered here more
reads come here likewise a promoter and so on So this is the way depending on what you
are looking at transcription factor and others you are able to infer as to how that could
possibly you know regulate the expression of the gene So that is the power of the next generation
sequencing and and as you have seen the just you know these technologies are coming up
still you have long way to go and maybe next 5 years 10 years you are going to see you
know changes that that you cannot imagine beyond imagination and it could be anywhere
in the city town there could be shop that can sequence the genome and may have methods
to analyze the genome and predict as to whatever risk factors that you have or diagnosis that
can you know become better just like what you have for blood tests and others So with that we end this section of the discussion
that is next generation sequencing The next classes we are going to discuss about the
usage of these sequences to understand how we evolved and how we can analyze the genome
sequencing

Leave a Reply

Your email address will not be published. Required fields are marked *