Speech Transcript - UNCOVER COOPERATIVE GENE REGULATIONS BY MICRORNAS AND TRANSCRIPTION FACTORS IN GLIOBLASTOMA USING A NONNEGATIVE HYBRID FACTOR MODEL

0:00:13	i can give a
0:00:14	talk
0:00:15	so that uh
0:00:17	where your your stay
0:00:19	so
0:00:20	i think a my job of the easier a because uh a a a lot of stuff and of the
0:00:24	background and and all those
0:00:25	actually actually introduced by
0:00:27	there is talks
0:00:28	so of what what we're trying to do here is a to uncover the
0:00:32	to operate of regulation by
0:00:34	transcription factors and michael R As uh using a bayesian uh
0:00:39	basic it's uh
0:00:40	this is it's a regression fact mall
0:00:42	or or call this so hyper affect them all
0:00:44	so
0:00:45	uh what's the object of okay uh the objective is saw
0:00:49	to understand how gene expression basically transcription this being regulated by
0:00:53	transcription factor all this common knowledge at
0:00:56	and my car it a it's a small molecule that's side recently
0:01:00	oh uncovered to normal also regulate a transcription
0:01:03	so
0:01:04	what what this approach are a so basing i wanna come you that that we can use this saw a
0:01:08	base fact factor model to to to serve the display
0:01:11	per
0:01:11	and um using the small on top of uh
0:01:14	michael or a expression data a lot of biological a prior knowledge
0:01:19	so
0:01:19	a just a little bit of background or which are sort already been introduced by a lot of uh uh
0:01:24	a previous of speakers
0:01:25	so this is essential that number like biology uh
0:01:28	so it's uh
0:01:29	i it to say uh that the you you know more the information flow goes from a D and they
0:01:34	am are eight uh in and the protein protein use a basic building block all
0:01:38	all living cells
0:01:40	uh so here you know my focus is on transcription so basically how D and they it's been transcribed into
0:01:45	M R
0:01:46	uh a this process so here looks lean your but actually it's being
0:01:50	have really regulate it okay and a male it's rewrite the by two factors or the first one is a
0:01:56	proteins call transcription factor so now you're looking at the D and a okay the transcription basically is a copying
0:02:02	of one change in the D and they
0:02:04	into the small molecule into the small coke M are a and then by M R you being later translate
0:02:10	like to to protein
0:02:11	so that the rank of the first regulator their is called transcription factor in a lines to they was up
0:02:16	from or region of the ageing source for example this is a gene
0:02:20	and then the controls the
0:02:21	of the product of uh or expression sure
0:02:24	oh the M R a
0:02:25	and for their know recently people also understand that the
0:02:29	uh and that the small molecule actually is come my core
0:02:32	it's a no i i rolled a soft and i are it's a little bit confusing with mike
0:02:37	it actually binds in that so called mold the region of the E and that the search to be great
0:02:43	degree
0:02:44	M are so act actually no
0:02:46	a together uh the transcription fact and Y core it together actually it's
0:02:50	a kind of a better explain the complexity of of of the leaving so why we have this type versa
0:02:55	a a a a traditional way if we want to look at transcription factor we we pretty much have a
0:02:59	similar set of a transcription for
0:03:01	we're difference of white
0:03:02	so my or actually give you another lady of explanation
0:03:05	so here might per mike my goal is okay okay a right now on you we have to my are
0:03:10	rate high so see in that can be used to
0:03:13	we the measure
0:03:14	mri more case so
0:03:16	for example we have michael or rate which also be introduced by a of three speaker
0:03:20	and also
0:03:21	well i happen to be that the we can also since this is also a
0:03:24	are in a week it was a measure michael
0:03:27	so
0:03:27	we we are the goal here is to
0:03:29	to really understand how am are watching gene transfer transcription be regulated by
0:03:36	my car and transcription factor based on
0:03:39	and M a measure all my mike or measurement this but there a case and michael R
0:03:43	mission
0:03:44	so
0:03:45	oh before supposed to be a like in
0:03:47	in which he yellow but now it's black
0:03:49	so basically that's the that's the goal here
0:03:51	or rate uh so that's
0:03:53	these to two factors okay so let's see what are the calm approach has been taken all
0:03:57	so a basic this clutching not work or in a a a a very simple way so you probably see
0:04:03	these things how how i don't often right
0:04:05	so it it's it's a very messy network well normal each not represent a gene and then you have links
0:04:11	score L linking these different genes
0:04:13	so
0:04:13	uh how we only are the meeting here is a pretty much a that's okay if two genes a link
0:04:19	they like to are sorted
0:04:20	but
0:04:21	but the problem here's how do we interpret a especially if i one to understand transcription regulation
0:04:26	or what does this king really
0:04:28	tells about transcription regulation
0:04:30	oh oh it's very very difficult actually because all these only says
0:04:34	gene are also she don't say whether
0:04:36	the ching a association through the transcription regulation or some thing
0:04:41	uh interact also also forth
0:04:43	so this is actually a
0:04:45	a a a i be working on this before about that you a kind of a stay away because it's
0:04:49	so hard interpret and you when you present a just they they don't know what
0:04:53	i don't even know what to tell them and they don't know how to interpret
0:04:56	so basically i one you know look at them more D to about of biology and see if whether we
0:05:00	can really model of this process of a transcription regulation and my car regulation
0:05:05	oh oh oh star by modeling like a everybody that's here so we assume that the transcription fact is a
0:05:11	protein so so that the protein activity we call this a acts a so the actively that and you know
0:05:17	the the
0:05:18	or
0:05:19	a all basic basically a little bit of a about on the a basic on it the more transcription factor
0:05:25	a you have prop possible in little wreck a so are quite a bit of a a a a a
0:05:28	gene transcription
0:05:29	so we use a to represent a transcription factor
0:05:33	putting level activity and then use using Z
0:05:36	to
0:05:37	oh denote the
0:05:39	expression level of
0:05:41	michael or are okay and then where we're saying seen
0:05:44	my car and transcription factor regulates the gene product are which is a are eight where we call why
0:05:50	which can be measured by the mike or read data and and
0:05:53	here we call what and then we can relate to this relation by a a simple linear relationship a where
0:05:59	we stay okay uh the R a expression level as
0:06:02	do to
0:06:03	on the regulation or the bond that is a axis actively putting of activity of transcription factor for at
0:06:10	and
0:06:10	the mike or expression level a K and a and B U are the so so called with three
0:06:15	or coefficient
0:06:17	also this this very simple model and this is a free much just that's okay but in the case model
0:06:22	the case where there's a the one shows can fact and the one my car a reality actually is a
0:06:26	lot or more complex where you ball a lot of my car a lot of can with fact
0:06:31	oh uh and and again you you're gonna have a more on that phone it
0:06:35	a model like those for one
0:06:37	E
0:06:38	and the if you really although the in higher G know where you have possible week one T uh to
0:06:43	forty thousand G
0:06:44	and then you you're looking at basically a matrix like that in this but you're case the measurement R expression
0:06:50	ah
0:06:51	uh this is a matrix for each will represent aging and each column vectors represent
0:06:55	he sample for competition the patient soul also for time
0:06:58	points
0:06:59	and the X the that it was here you know well represent expression mri of one
0:07:04	so i assume early acts here represent a
0:07:08	that's cheating that's sample and and an is the transcript factor of this act
0:07:13	i and does Z as the mike or a a michael are and they
0:07:17	activity sorry this is so this wrong
0:07:19	and where i i slice said before this can be measured okay this can be measured by
0:07:23	mike or all of high simple sequence
0:07:25	and at
0:07:26	S the so called a three stress and the B as in my regular wrist stress a and E is
0:07:32	the ad it to i it to uh at
0:07:34	what
0:07:35	or right so basically we're we're looking at the C creation now uh we're given Y and Z E we
0:07:40	tried to
0:07:41	a secure rubber some white and Z base on this model
0:07:44	and uh
0:07:45	and uh so this is a goals of data is given Y Z what one understand
0:07:49	a B and
0:07:51	i
0:07:51	a so how do we how how are we gonna really achieve this
0:07:54	a so
0:07:55	traditionally additional the of just have a model is really a a a a factor regression model this part is
0:08:00	the fact the of this part as a regression model a so this nothing you are and you you know
0:08:05	and and the solution you can see a a couple of different solutions pca i C S already be to
0:08:09	use by
0:08:10	but make less and uh
0:08:11	and it and M have a a row one are good at this type of the mall or not really
0:08:16	sufficient to to really model
0:08:18	the D to while the white
0:08:19	so the reason i give you a very simple reason here for example in you are we're looking at a
0:08:23	this is a relatively uh a real scenario you like you can kind of a
0:08:27	and get a sense okay
0:08:28	a so
0:08:29	yeah if if you want to use pca to kind of a although does a basic P Z sense of
0:08:34	the loading matrix for this this is a a a more make be an a somali matrix
0:08:39	so well we make must for right okay i believe all it
0:08:42	are are very you know and now we know that each gene transfer or fact actually regulate only a very
0:08:46	small set of genes okay
0:08:48	while relative to okay it's all what a couple of
0:08:51	yeah up two thousand a couple thousand genes
0:08:53	still in in terms of the overall number of genes which is twenty thousand to forty
0:08:57	as a sparse hiding
0:08:59	so of the major should be spot
0:09:01	and also on you you know where you have a regulation where or is it now as your abdomen
0:09:06	can be we have we already accumulated a lot of are not just to which you know transcript fact to
0:09:11	regular what's set of a
0:09:13	so we should be able to you incorporate this type
0:09:15	oh not
0:09:16	and thirdly
0:09:17	or so these samples actually an you know you look at the sample
0:09:21	you like a sample these samples but like a whole you are represent for example patience you know the patient
0:09:27	measure
0:09:28	and saw in the case but these disease
0:09:30	uh some some patience actually have similar
0:09:33	expression path
0:09:34	and meaning that they they have these can be used to define
0:09:37	the stop type of disease
0:09:39	also if you have most similar stop five
0:09:41	you're expression level should be it
0:09:43	so these problems are i actually should be carly
0:09:45	to re represent the condo
0:09:47	so something like
0:09:49	i start X and Z a get this from a factor
0:09:52	activity as michael
0:09:54	or you should have a saw
0:09:55	correlations you should have these group
0:09:57	a in the set
0:09:58	and what in C uh it doesn't really models
0:10:01	it like this
0:10:02	and also a lot the transcription transcription factor activity should be known that
0:10:07	like a what the uh make and also argue
0:10:09	a but that that was in the case of the gene but there's similar market
0:10:13	so uh we had we need to model really non negative a transcript five
0:10:17	i
0:10:18	well
0:10:18	in the case of my car in my car known to down regulate the transcription
0:10:22	so it's loading matrix must be
0:10:24	negative
0:10:25	or a loss of this matrix
0:10:27	actually is used to be negative
0:10:28	so we need to somehow in all these T to by all the in to the ball that you know
0:10:32	to do
0:10:34	a a a basic and i'm gonna tell you how we we we model all these
0:10:37	each of one of i
0:10:39	they
0:10:40	all
0:10:41	or to start with a a a a sort of basic the modeling goal was to model the sparsity
0:10:46	a and B or in knowledge and uh
0:10:48	yeah a a model the non-negative transcription fact
0:10:51	video
0:10:52	and a
0:10:53	negative regulation my car
0:10:55	and then
0:10:56	a the sample correlation
0:10:57	okay so you need a lot of things
0:10:59	a small
0:11:00	the start with the sparse that we use exact the same model as that is what uh
0:11:04	the close on johnny
0:11:06	uh
0:11:06	and uh L actual was using here
0:11:09	yeah yeah the spy high just one a point out actually use high right of basically notes over bit
0:11:16	a probability of transcribed factor L regulating gene and so this can be really them
0:11:20	there's a lot of prior knowledge available from that
0:11:23	they
0:11:23	so we can really incorporate
0:11:25	these prior knowledge
0:11:26	a in two
0:11:27	a time
0:11:28	and a a so this is how we model the sparsity of a a a well in the case of
0:11:32	a
0:11:33	B actually very similar model yeah
0:11:35	here
0:11:36	now we have to use it a gaussian
0:11:38	to really out of the down regulation of a a of a mike
0:11:42	a B as the regular matrix um mike
0:11:44	so that's only differs and i
0:11:46	again as a prior knowledge and there are all their a databases
0:11:50	and also
0:11:50	part
0:11:51	well
0:11:52	as a oh
0:11:53	yeah is just to a point of a like card regulation is do a very active or research just so
0:11:58	we don't really know exactly how my or a
0:12:01	right right of the genes not at the level of transcription fact yet
0:12:05	but are
0:12:05	a target prediction out
0:12:07	that can be used to really a a give you some prior knowledge here
0:12:10	so that's how a model the sparsity and copy the part
0:12:14	but apply
0:12:15	then let's more want to
0:12:16	a a needs to be non-negative transcript factor i
0:12:19	a body
0:12:20	so it's not using actually trying to go also use the right
0:12:23	only differences
0:12:24	we have a mat
0:12:25	actually and zero
0:12:27	this possible i
0:12:28	yeah you know there two
0:12:30	of using rectified of one is
0:12:32	it introduces a
0:12:34	additional sparse the actually even a transcription factor activity
0:12:38	and also a it gives a very nice the a function
0:12:41	uh uh formation for the base and uh
0:12:44	uh i
0:12:45	base and duration so that's how i
0:12:48	um
0:12:49	fact
0:12:50	and then
0:12:50	owing to the correlation sample correlation of be fine example stuff
0:12:54	so patients
0:12:56	well we use basic assumption is that he's
0:12:58	yeah samples are the same ballpark are so it's a natural plaster model
0:13:03	so mixture gaussian
0:13:04	and
0:13:04	a problem with mixture girls
0:13:06	you know that the fosters so we actually use of duration should process of mixture
0:13:10	now
0:13:11	a
0:13:12	i
0:13:13	sure
0:13:13	or
0:13:14	do should process of make sure a rectify
0:13:17	record
0:13:18	what we
0:13:19	in use a duration process
0:13:21	yeah
0:13:21	so
0:13:22	putting a in everything together this pretty much uh what the model looks like a also we have all these
0:13:27	different parts we
0:13:28	a basic a for you do the factor
0:13:31	right uh i not
0:13:32	factor not
0:13:33	or projection
0:13:34	and uh if you put in of them all
0:13:37	looks like
0:13:38	i
0:13:38	so a lot of parameters to estimate at the of a sure yeah every why
0:13:43	and then that the so how have
0:13:45	resort to some of
0:13:47	you you a traditional
0:13:49	something for a for example
0:13:51	you something
0:13:51	for in this case because of these very powerful
0:13:55	uh which also prior distributions
0:13:57	we have thought conditional distributions in close
0:14:00	oh
0:14:00	base this uh but uh
0:14:02	if sample thing of what am i'm not gonna do on the on the durations is along long
0:14:08	i
0:14:08	i
0:14:09	but
0:14:10	wise as they we are able to really create this
0:14:12	beep something solution
0:14:14	so was start by looking at the a like assimilation data
0:14:17	where
0:14:17	uh this but
0:14:18	a we have one fund
0:14:19	genes
0:14:20	a with
0:14:21	well some of us
0:14:23	or are are this
0:14:24	a this is called
0:14:26	how a most rate with the a real situation
0:14:30	oh
0:14:31	later
0:14:31	it's
0:14:32	so and we assume there flight faster
0:14:35	and
0:14:36	and that are they a fully uh thirty five seven
0:14:39	are we look at uh
0:14:40	basic
0:14:42	as
0:14:42	are
0:14:43	a a wall here i and talk about the correlations are with real
0:14:47	also
0:14:47	and i E uh
0:14:49	mean square error estimate a were also look at a sparsity you
0:14:53	station
0:14:53	oh
0:14:54	and the class triple form
0:14:55	so
0:14:56	just a an idea of how
0:14:58	samples are were
0:15:00	as there a case
0:15:01	actually rather to the fast ball of course this
0:15:04	and
0:15:05	a a high any on what what kind of a different the settings and error
0:15:09	so for
0:15:10	but uh generally it become verge are relatively that
0:15:13	a so this is the actual a be the cluster id so in this case was the two clusters
0:15:19	uh you can see it actually covers
0:15:21	a fairly
0:15:23	fast
0:15:23	so this is a and you know we we actually
0:15:25	look at a performance
0:15:27	a for different noise conditions
0:15:29	in of it
0:15:30	moist of errors
0:15:31	and for example in this case
0:15:33	why i we look at a this is a the this the so can estimate of a non negative or
0:15:38	i
0:15:39	spar
0:15:40	as far matrix and we look at the precision and uh
0:15:43	when the noise actually increases
0:15:44	oh of the precision actually a
0:15:47	when the boys
0:15:48	increases i
0:15:49	precision actor goes
0:15:50	well
0:15:51	i it some are but
0:15:53	a and then both goes down this give case
0:15:56	and uh but uh if you look at the faster actually class simple form a rates uh with the
0:16:01	and also the estimation
0:16:03	a with increase of the noise
0:16:05	and then we look at the data base of because
0:16:07	for for knowledge and the database has problems
0:16:09	so
0:16:10	a a a a there two type of problems for example whether the database really for all the norm knowledge
0:16:15	like a like what of the database
0:16:17	the whatever of data we
0:16:19	oh
0:16:20	for you know a precision again the precision recall problem
0:16:23	we set up a a precision of the data and look at again you know that you better performance
0:16:28	oh what what is that can be seen here data is precision
0:16:32	you know increases
0:16:33	you know if whatever big report basis
0:16:36	to and we be able to really recover
0:16:38	i
0:16:39	well
0:16:40	these on uh
0:16:41	S regulations
0:16:43	um nonzero out
0:16:45	uh
0:16:45	so
0:16:46	uh
0:16:47	i i i can speak this uh this uh
0:16:51	one
0:16:52	job
0:16:53	right into the real
0:16:56	a real data actually were looking at that we using the the
0:17:00	a cancer genome at
0:17:02	yeah
0:17:02	this is
0:17:03	in H
0:17:04	a project
0:17:05	and we take a look at the meal
0:17:06	a
0:17:08	right
0:17:09	a
0:17:09	oh where and you know particular we we look at a a of a form
0:17:13	haitian
0:17:14	i
0:17:15	yeah a gene expression data
0:17:17	and then a about
0:17:19	one patient
0:17:20	i
0:17:22	i
0:17:24	i
0:17:25	oh
0:17:25	yeah
0:17:26	uh what what we and we need to also have thing
0:17:30	i
0:17:31	and really look at a on their own the show what
0:17:34	perdition
0:17:36	oh
0:17:37	but
0:17:38	yeah you know just okay
0:17:39	extract now
0:17:40	that
0:17:40	all these conditions having my
0:17:43	or
0:17:51	or
0:17:51	actually patient
0:17:52	samples
0:17:53	in addition
0:17:55	one norm
0:17:58	we have forty that
0:18:00	i
0:18:02	yeah
0:18:02	we go the original why
0:18:05	just to ask
0:18:07	or what are
0:18:14	a
0:18:15	i
0:18:16	yeah
0:18:17	i
0:18:18	and
0:18:19	i saw that my
0:18:21	first fine
0:18:22	so which in fact
0:18:23	yeah seven michael
0:18:25	okay
0:18:26	and i
0:18:26	so with this are we told to be a basic the database P and C one are the try change
0:18:32	step possible regulate
0:18:34	i
0:18:35	these set the transcript fact a seven my car
0:18:37	and uh also in uh we a come up with a hundred thirty five genes
0:18:41	so
0:18:42	uh so
0:18:43	the supplies to say that these a hundred thirty find genes
0:18:46	are regularly by D
0:18:48	all of these are set the my car some transfer factors
0:18:51	in
0:18:52	in many maybe to the conditions because all of these these uh prior knowledge are are are are are derive
0:18:57	from an a different conditions
0:18:58	but they are not necessary true but you heard too
0:19:01	post
0:19:02	one
0:19:03	okay
0:19:03	so and then for a a as to the prior knowledge for a transcription factor
0:19:07	regulation way go to be trends back a and then extract these a regular regular three
0:19:13	a a prior knowledge and from like a regulation actually we have our in house uh
0:19:17	prediction uh we
0:19:19	a these two papers
0:19:21	so that's all these set all the of the uh basically a the experiments
0:19:25	and then this is the
0:19:27	uh i for the of proper ability poster their probability of nonzero elements and thus against can see most of
0:19:33	them are nonzero or the probability of a
0:19:35	one want
0:19:36	it's very very small
0:19:37	and only a small set of a probability actually give you
0:19:40	close to one
0:19:41	so and then all
0:19:43	a not all these possible links we uncover a one hundred uh
0:19:47	or regulations as so this is a side sparse
0:19:50	and and into wrestling eh
0:19:52	so uh one look at the covered regulations there are about the uh one fourteen i read report in the
0:19:58	database
0:19:59	and eleven
0:20:00	a
0:20:00	in the database are not really on cover a uh but we pick up seven additional you are regulations
0:20:06	which are not really or in the data
0:20:08	and then this is the the so regulatory what
0:20:11	and this actually each node on the site represent a a transcription factor each no on this i represent my
0:20:17	car and the circle here there are small those act actually stacked together and their are the represent right they
0:20:22	represent genes
0:20:24	and each link it has to very clear interpretation
0:20:27	so have a here pretty much it means that this in fact the right that that that change
0:20:32	and also
0:20:33	uh we can use the the loading matrix the as a loading matrix
0:20:37	to to to the to to uh to
0:20:40	indicate whether this regulation of transcription factors up regulation or down regulation
0:20:44	well for my car is always that regulation and this is a a a a heat map of a the
0:20:49	loading matrix as so there are a lot of zeros basically here
0:20:52	oh so
0:20:53	and this is a cover the uh the transcriber fact activity all the fact
0:20:58	models and together with the measurement of the my car but this is the for change remember i tell you
0:21:04	in the sample there's and there's a normal some so we basically use an more samples of control
0:21:08	to calculate the full change otherwise transfer factor
0:21:11	oh about should be all
0:21:13	cost okay and then this is the the cost or that that's be uncovered
0:21:17	by the model so basic it model un covers three cluster
0:21:21	and then also it it's a see the expression levels are more less the same within in the cluster
0:21:26	and then we look at the saliva for each group of but these cost
0:21:29	see whether bit do of form some trouble socks of a sub group a sub D C stuff sub type
0:21:35	answer
0:21:35	so the we we look at is so we look at the a bible with the see whether after treatment
0:21:39	that the the the the patient in the same group have a similar so bible
0:21:43	so
0:21:44	it's a seeing you all um they
0:21:46	the different groups spatial in different groups
0:21:49	in have a difference some why will meaning that that you this separation does been something okay can indicate basically
0:21:55	you
0:21:56	an next um maybe see a patient this be can say that this patient possible after after treatment and need
0:22:01	longer that the patient in
0:22:02	this point group okay
0:22:04	and then we look at the the basic the P of the pure voice
0:22:08	i these source some bibles
0:22:09	and the
0:22:10	and the a clearly shows the who uh the the a cost or one the faster
0:22:14	to actually has to large as as a what different
0:22:17	so
0:22:17	they they can be really are used to
0:22:20	in as a viable of the effect effect is up a tree
0:22:24	and uh so we were going back actually going back to the mike car an expression data C what thing
0:22:28	you know you can
0:22:29	a come up with a similar result
0:22:31	ah
0:22:32	by simply using by or and data and my a and the G did come by my car inching get
0:22:36	a without going through
0:22:38	the the factor analysis
0:22:39	just basically
0:22:40	for for one class room uh direct on these
0:22:43	individual data
0:22:44	and the reason uh the that can in is no i this is these are the P about was actually
0:22:49	this is our perform this a lot log P about the as so we have a
0:22:54	signal actually higher P about then if you
0:22:56	look at my car
0:22:58	gene or my car engine together a lot a without using the factor analysis
0:23:02	so this really shows that you fact
0:23:04	effect fact than this all of this
0:23:06	fact model okay
0:23:08	uh so that pretty much come close my my also
0:23:11	keep those that uh
0:23:12	just could that the in
0:23:14	a a from a a in a H and uh
0:23:17	and uh
0:23:17	uh uh that okay
0:23:19	thank you
0:23:20	thank you i
0:23:24	what what two questions
0:23:27	yes and yes
0:23:32	oh your examples
0:23:34	you have
0:23:35	quite
0:23:36	small number of genes is
0:23:38	uh i
0:23:39	it is the factor analysis that that that you uh you have uh
0:23:43	a restricted to
0:23:44	very small sample
0:23:46	so far yeah
0:23:47	yeah
0:23:51	i had it to use it T question a mode to model can you go back short
0:24:02	each each one
0:24:04	the matrix
0:24:05	so the form
0:24:06	yes this one is uh or
0:24:09	this one yeah
0:24:11	so you then you i don't if you but problem you mean the of course uh if you if you
0:24:16	are one in C can you find the unique solution for a X in B
0:24:21	of course for E X to me know do that
0:24:24	the uh yeah
0:24:25	yeah a very can question are
0:24:27	actually the something really worse than a starting here we like i can give you a radical
0:24:32	you know a cool whether there's so
0:24:34	i by with it
0:24:36	or
0:24:36	so actually uh additional of things i haven't really talk about a for example no we have to restrict that
0:24:42	a
0:24:42	the
0:24:43	the call uh of the uh
0:24:46	all
0:24:47	the factor needs to have a a a a a unique there
0:24:50	and uh
0:24:51	and also of the columns of uh
0:24:53	oh a case
0:24:55	these to have a
0:24:56	a the same
0:24:57	where it's actually
0:24:58	the car
0:24:59	and also
0:25:00	uh
0:25:01	the the the hours of a at and Z should be so be we we have to do some three
0:25:06	prof
0:25:07	make that is a a and C to be
0:25:09	in the
0:25:10	same Q
0:25:11	you know as you have a
0:25:12	i can't define all a and B
0:25:14	okay but a
0:25:15	whether there
0:25:17	what what is the competition i can't
0:25:19	that i i i i can at
0:25:23	okay okay thank you things
0:25:25	for at the end
0:25:27	and uh i i you back

UNCOVER COOPERATIVE GENE REGULATIONS BY MICRORNAS AND TRANSCRIPTION FACTORS IN GLIOBLASTOMA USING A NONNEGATIVE HYBRID FACTOR MODEL

Systems Biology