DevoWorm Lab Meetings, 4 Jul 2023

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: DevoWorm (2023, Meeting #25): DevoNet, TDA/GNNs, Open Biophysics Problems, and Embryoid Models

Description

GSoC Updates: Devonet and model selection, topological cell segmentation. Discussion on Topological Data Analysis (TDA) and GNNs. Large Language Models for genomics and HyenaDNA. Discussion on biomedical image sampling problem and creating biophysics graphs. Embryoids III (more on human embryoids). Attendees: Sushmanth Reddy Mereddy, Jiahang Li, Bradly Alicea, Himanshu Chougule, Susan Crawford-Young, Richard Gordon, and Jesse Parent.

A

Hello, how are you guys, nope hello? How are you Manchu yeah, oh Susan,.

B

A

All right uh so welcome to the meeting uh that looks like we want to do. Updates uh I'll do such in order and then I, don't know if Susan has any updates, but then we'll get to her.

B

My graph is still upside down. Oh.

A

C

Oh can I start my update, then yeah go ahead. Yeah last week, I completely worked on the model architecture. Actually I started.

C

Yeah uh there are two photo library: four folders here: one is Library folder, where whole code is their model architecture, how we will take the exit and all last week, I worked on the model and the loss function uh here. I have took two modems for nuclear segmentation Network one modem is there new phase detection Network one model is and yeah uh it's completely written in by torch.

C

I made some changes according to what I needed and the last function I thought of using the direct line, but I thought a single line from pipe, but I thought it wouldn't be that appropriate. So I wrote my class for own dice class function from uh net resources and I'm, giving I'm working on that. Only and uh last week I wrote last four week blocks all group yeah.

C

Can you see this thing? Is.

A

uh It doesn't look like it. You gotta, reshare, Maybe,.

C

Yeah yeah I first I wrote the whole code here and below that explanation of it is that if someone comes up and sees our development code, some may may not understand new beginners, so I thought of writing a Blog about it. Every quote: I wrote so whatever I am writing those I am just explaining you to make clear way in the points whatever each word means in that. So that's what I am doing uh if some I was trying to comment the whole code also, but right now.

C

This is my process like in this week for I worked on the data. So it's fine 3D data sets right, I wrote in every explanation of it and then I mean this last function. Also I don't report it, and this week I completely worked on the model architecture, so my main motive would be writing uh code. Explainable code of model architecture, yeah I, have done these things only this week and I tried to work on Sam also but I couldn't handle both at the same time.

C

This week, I will try to manage model architecture with sound model. Also by next week, yeah I'm just working on the model right now uh there are no fine errors. I am using ciprit library to image, processing, so image processing, because Mayo once told me that he wants to use circuit image library for develop, but he couldn't use it of last time, so I started to use a scientific image library for the level.

C

Naturally, it is mentioned in development issues phase also that psychic image like update to psychic image Library, so whatever I was working on I, am writing. In psychic image, only I am using a special method called Watershed segmentation. uh uh It generally helps like when two cells are overlapped to each other. It doesn't serve it as a single cell. It takes us two different cells, so it would be more clear to us for finding the centrants of each cell rather than overlap.

C

So that's what my thing is going on. I am updating it to Mario calls already watching the whole project. All.

A

C

I mean Bradley if you want to take a look on the blocks. Also, please take a look at because I try to explain every word in a clear way that if someone reads it, they could understand it before.

C

C

And yeah: that's what my update is: I completely worked on the model architecture. If you see here, you can see.

D

C

Yeah you can see here one this part is for NSM model N2.

C

Yeah, this part is for NSM I, usually mainly convolutional 3D, because whatever data set, we have this 3D data set. So I used a 3D convolutional layers and tried to implement it, and this part is of unit I just want to compare it with uh our Network and unit Network, so I'm just trying to implement it. Mostly I will remove this part of Port, because my proposal was made upon NSN and Indian. Only if I can compare these results of unit NSN NDM I will try to implement it and see.

C

First I was working on this. No, no! It's not similar to unit in 3D I was just comparing the segmentation of all these three networks, Sam model and whatever I proposed in the proposal and unit model, which are gives the best results. That would be useful to us. I'm just checking on which segmentation network works. Fine, okay,.

A

So you're just making a comparison between the two types of models, so yeah yeah.

C

Just proposing my model wouldn't be fine right, proposing comparing with other models and if my model performs well I can just mention it. Like my model. Power perform with these models.

A

Do you could you describe the uh what what a unit model is.

C

uh Unit model is mainly used for medical image processing. It has uh nine convolutional layers in a u-shape. Let me show you with an example.

C

Yeah, this is how it looks in the shape of I mean uh there are nine convolutional layers. First, continuation will happen after that. Deconvolution will happen. You're.

A

Not sharing your screen again, yeah.

C

I'm not sharing your screen, that's okay, sorry yeah! These are nine convolutional layers totally first convolutional convolutional will happen until here, then decodational will happen, and this is mainly used for medical image processing. See here. You can see biographically image segmentation yeah, and this is the old model I'm just con, seeing which gives a better result either Sam model or demonet model or unit I will try to extract I will take two I will try to compare it with the rcmn model, which was implemented by himanshu.

C

Also I will just compare the results and provide the graph which model outperforms, which on which one performs well I, will take it as for better one. That's what my idea is just not giving up the model. I am telling it looks good, but with the perfect explanation which model out corresponds with and according to situation, we can use the model.

A

Yeah, that's good yeah that should be interesting to see which ones perform, though.

D

You're going to use like a benchmark.

A

Data set, or what's your, what kind of data center you're going to use for this I'm.

C

Using cell tracking challenge and with that I will I will just take a common I will eliminate some part. Some images from the cell tracking and I'll just check the test results of all models on that and after that, I will produce the loss, function and I I O, U score and I'm just trying to compare them. If the io user and the loss we can just find it out which model performs well.

A

Okay, yeah that'd be interesting: okay, yeah, you Nets are yeah, I mean they're, probably older, relatively speaking, but they're also specialized for image processing. So.

C

Yeah, they are mainly for biomedical segmentation only. They are mostly used in some segmentation or concept implementation, so Etc they are particularly yeah. I mean microscopy images, for that only you know.

A

Yeah, that's great, thank you for that update. So this week, you're going to be working on that comparison and getting the.

C

Yeah I was gonna, I was thinking to work on the training, the model because the model is not completed yet I tried to write the model architecture, so many errors I have encountered, but at last week is working fine. This week, I will write the trainer code for it and yeah after that, I will try to train the models. Actually I was having GPU issues. uh All kaggle gpus are completed, which I have so I was I took a I buyed some of gpus from online from Google collab of.

C

A

Good yeah, that's always a good resource. School lab uh for doing that sort of stuff.

C

Yeah, actually, the resources are emptied because uh collab provides up till like around I, don't know 50 GPU, compute units I, think so I, don't remember, I completely used them off for this month and I use a kaggle also, and the work has been stopped since two days, so I took a another resources: I bought it from I, bought it 100 compute units I'll, try I will try to try in the model. That's it.

A

Sounds good, thank you, so hello, dick and Gia Hong and Jesse.

A

Come on shoe, are you ready for an update.

E

uh Yeah all right, yep, okay,.

E

uh Is it visible? Yes,.

A

I can see here, Okay.

E

So uh I was working on the model data I initially proposed about the mask CNN model and I was able to like figure out how uh how to solve all the errors and uh basically create my own custom uh class for the segmentation data set in which I could incorporate like the image mask, as well as the bounding boxes for it.

E

And after like a few error, solving methods I was able to like finish this uh class of the data set which just Returns the transform damage, as well as the Target and the target has all the information which is needed for like segmentation.

E

Let us the boxes, the masks and the area of the Boxes Etc, so I was able to uh use this and uh like uh successfully uh get the data set uh Bible for training and then I basically uh created the model for like uh instant segmentation as well as uh I was able to like use the bounding, Boxes Etc and also create visualizations for it.

E

So this is how uh it's it's it looks and uh for the data set uh that I've created so and now, I also trained it for like uh just I used like the basic optimizers and schedulers, and I tried to train it for like two ebooks and like each Epoch took like around 35 to 40 minutes because, like the size of the data set, is a lot large and and then I just saved that model, and uh basically since I, just since uh I was just checking if I could train it and if the model fits perfectly so I just standard for two ebooks just to see that and I was able to uh like this is like the bounding boxes and for the prediction so, uh but, like the actual masks, have only got like only one mask as you can see over here, because it was just straightforter epochs.

E

So what my plan right now is to do like some hyper parameter tuning, as well as, uh like, uh probably like make some changes in the data set but like if I'm creates, making some changes. I'll like uh talk with sushma as well, because he's also talking uh figuring out how how to deal with the big data set and all and similar to him.

E

I, also why I have zero complete units available and uh which is why I was thinking of doing like some kind of parameter tuning uh on the CPU as well, but like after uh done making some small small changes in the model. I'll just uh get some Computer Resources and just finish. The training like on one night session like just keep my laptop running for like 10 hours, I guess, yeah, yeah and yeah I've not used tackled uh until now.

E

So I can just do the hyper parameter tuning on toggle and then I'll drive it back to collab and, along with this, I started like uh working on the demo. This demograph part of the project like to derive or find a better ways to derive graph embeddings and right now, I'm, just in the uh learning stage of it. So I've been I.

E

Also watch the uh like the recording, which was given uh of the video like in which Zhang like explained uh like what is uh the way of doing the preferential attachment and also I'm doing that simultaneously as well.

E

So uh the data set I've been working on is also the cell tracking Challenge, and uh after uh this month uh after, like customize pipeline is created, I can work on the two over the different data sets for the membrane segmentation uh and after that, I can just work on the data set, which I had proposed like two weeks back, which was like for the longest agent, which was like kind of elongated like this, so just to see how the model kind of, if it works on that or some changes have to be made etcetera.

E

So yep, that's my updates.

A

So that's good yeah. Thank you. um Yeah I heard you met with uh with RV Raj uh he's the student in working.

C

E

A

In the orthogonal lab and.

E

A

Did that meeting go.

E

uh Yeah the meeting was great and he basically like kind of taught uh told me about the day. He wanted the things to go like to use Mesa, instead of like uh the which we initially did for asian-based modeling, that is, to use net logo and now and I. Think that's the right direction to go because, like uh using Mesa, is different from net logo, because that logo has its own architecture, and you have to follow some rules that that logo has for it.

E

So uh creating the agent based models for that uh for that environment itself is different than uh what we can do in Mesa.

E

So, uh like I, came from a basic walkthrough of what he should be doing and what should be the next steps like to get some kind of a small uh algorithm of agent learning algorithm ready for like uh some for uh the reinforcement learning part of it so like you can just do a basic file iteration and then move on to Q learning, uh just for a basic implement creation of it and figure out a way to just connect those both things like the Asian Taste algorithm, as well as the environmental Creator.

E

So that's where and also resources that I've used myself for are learning okay,.

A

Yeah, thank you. We were discussing this uh Friday about uh kind of the plan forward and he's also talking to ancient Grover who works in the lab, so he's acting. Kid is actually really interested in RL. So um you know that's that's going to be fun to kind of develop the RL capabilities there um yeah. So that's great!

A

What are your plans for this coming week? What do you plan to do.

E

uh So right now I saw a nuclear segmentation graph as well and in the meantime, I'll also look into what the open sustainability project is going. So I'll try to attend the meat. On fact. Oh.

A

Yeah that'd be great yeah, so you know sushma has been at the Friday meeting, so we run a Friday meeting it's at uh noon, eastern time in North America. So we do. um uh We talk about uh the two projects in the orthogonal lab, but we also, you know, can talk about the diva one projects if necessary uh and especially when, if you can't make this meeting or you know for whatever reason, but we also have some Crossover with the projects, um so yeah that'd be good.

A

So thank you, Hamachi for the update.

A

So I, don't know if uh we had uh Dia hung a couple weeks ago gave an update on his uh some things. He was thinking about and uh it was mentioned by himanshu. So I don't know if Tia hung had a follow-up to that or not. If he wanted to talk anymore about that or.

F

Yeah well hi, okay and.

A

You hear me yeah yeah,.

F

Great uh a long time to see how yeah, yeah, hi yeah, uh actually I'm kind of big, because I'm uh busy as preparing for my integrated confirmation reports, and it would be like after one month and I, have to present to some uh professors for my degree confirmation. So it could be a little bit challenged. So yeah I spent some time on this, so um maybe not much updates, because I didn't of commencement votes on those on the research project that I mentioned two weeks ago and I see someone mentioned that in this meeting today.

F

um So uh so, firstly, like like what I mentioned in two weeks ago, I said: I wanna uh apply the topological data analysis to the analysis of of ASO temporal topology uh within the cell development process, and it is a general uh like General topic and I I haven't like thought about any uh like model design or some more specific details about them about this. Talk this topic two weeks ago and for now I think uh there could be some some approaches to solve this task.

F

For example, the first of all is uh compared with other study static system for the cell development process. We have a topology which is dynamic. It is time depending that means uh we have. We can explore the topology uh of like the cell dependence, Network or something else.

F

For example, if we have a microscope video- and we have this video recording the cell development process, we can sample a lot of friends and for each frame we can sample a lot of cells like using some sales segmentation method and and then we have a point count, and we have some ways to. Maybe we don't have to build a network, we can apply the position of homology, which is to explore the topological information underline this.

F

This uh points Cloud it could be a it could be approach, but if we uh deliveries this approach, that means we have a static topology for each friend. But, however, uh this topology is dynamic.

F

So that means we need to find like some ways to uh to build this Dynamic topology like because, uh in my in my opinion, I think if we just start, if we just consider the topology of each friend and we do not consider the temporal connection between uh the tolerance, the topology across different Trends, it seems like we are not considering the Dynamics of this topology. So this will, uh if we do not consider these dynamics, that means well.

F

We are regarding the whole stuff, as just investigating the static topology of of a network that could be Trevor, like uh maybe I, didn't make myself career. Like then, give your example, for example, uh we have another. uh We have seen another paper in maybe uh one one year ago, which is a topological graph in your network and I'm, not sure.

F

If you also remember that paper- and in that paper, the authors proposed uh graph, your network to explore the topology within uh static Network, for example- you you have a graph which is not Dynamic and you apply a topological graph, your network, to explore the topology within this network, so uh I think in. In our case uh it. The the approach that I mentioned before is very similar to this topological problem Network.

F

So, uh in my opinion, I think uh so. First of all, I can take some experiment to test how this topological graphing network uh performed in our case, and then I would like to try some the answers uh to export the Dynamics of this topology.

F

uh What I think this Dynamic topology is very very important, because uh you need to remember an example, which is the global rotation formed by the collective cell, migrations that I mentioned in uh two weeks ago, and actually, uh if you, if we want to like if we want to see this type of topology, which is a global rotation, uh most specifically a circle. If you want to see such real topology, we actually need to look at the spatial topology as well as a temporal topology.

F

Yes, so and I think for most of multicellular systems, then Dynamics is very important. So that means we need to find a way to investigate the dynamic topology, but first of all, I I. Think I would uh like try the technological graph in the network that the simple approach is for the static topology yeah they were about. They would be my first step and for the next step, I would like to investigate the dynamic topology like there could be a lot of matters to consider the dynamic topology.

F

The first approach is like uh you can think of that you for each frame. uh You have a state in topology, and maybe you would use some kind of embedding or some Vector to represent this static topology and then uh you can. We can like, like just like other recurring, that's work or other words which focus on the uh some serious data. Maybe we can use some recurrent neural network or we can use some Transformer to aggregate this to logical information across different fans. I mean this could be an approach and another approach is I.

F

D

F

More fancy is um uh because actually uh I, think persistent. Homology is a very cool, very Grace tour applied to the points cloud, data and I think if we have some ways to build such a points cloud, a close friends like a single point, Cloud improve uh like uh covering several friends. Maybe we can find some ways to use the positional homology to like investigate the spatial temporal topology I'm, not sure if I make myself career the the challenge here.

F

Is um you have you reading distance, which is the spatial distance and you maybe you can Define some other types of distance. There are other space resistance, but if we want to connect cells across different different frames, According to some some kind of reasons, that means we need to consider the the the 10 protists like the distance of course, time Dimension. It is not very intuitive, like the spatial distance like that, you can eat the distance. So this could be one trendy assume.

F

We need to consider uh how to define such a temporal distance, and the next challenge is how to unify the temporal distance and the spatial distance. So we can build a point Cloud. So we can apply the position of allows you to a certain point. Colleges. Yes, I think this this uh if they are reasonable or if they are meaningful, but maybe I would take a try after I finish, my information, yeah I, think.

A

F

That's right, yeah, oh.

A

Yeah, it's great yeah! Thank you for that. uh Yeah there we go there, we go thumbs up. um Yeah I mean that's. That I talked about some of that two weeks ago. uh The persistent homology thing, I, think, is really interesting from the standpoint of kind of taking these Point clouds and breaking them down into uh you know a network, but also into some of these higher order, shapes and things that might describe the data. So you know with with segmentation.

A

We have these Point clouds and midpoint clouds in time, but the point clouds in time are interesting because they're connected so like you have a point Cloud at time: zero and another Point Cloud at a time. One point cloud of time: two and you have cells dividing, so you have sort of this regeneration of the point Cloud as this maybe there's more uh increasingly dense structure with things in different positions, and then the idea would be that you want to link those things and we oftentimes.

A

We use um lineage trees to do that, but in a point Cloud, it's kind of hard to draw a lineage tree and then.

E

F

A

Where things moving in space, so we've you.

E

A

We've done some of that with uh well I mean we've done some things with different techniques. We don't have a really good technique for it, though that's the thing, and so that might be a good way to kind of build a nice model of that that change, but also the changes in position and all that and then characterize it as like this change in spatial orientation, or something like that.

A

D

A

You know this is like kind of a lesson of people working on Cell segmentation. That's where that go. That's where the data goes. Upstream in the pipeline, you go from like segmenting the cells to like analyzing. You know sort of a first order, analysis, which is the exploration of the point cloud. You can you know, label the point, Cloud points you know as cells and then you want to find the relations between the cells, and this is one of the things that we're trying.

A

This is why we kind of come up with these techniques um to kind of put together a model of the embryo, or at least the cells, and how they're sort of related to one another. So that's that's where this is going, and you know I like these kind of topological approaches, because you know I, think people have.

A

You know people kind of know, at least in C, elegans sort of what it looks like uh you know, in terms of you know where the cells are in space and where they're differentiating, they would kind of know how those things kind of unfold from the data. But we don't really have a higher level description of like what kinds of structures are, maybe precursors to tissues or what kinds of you know. Migration oriented things are emerging in the embryo.

E

A

Kinds of things we want to maybe describe with these topological techniques.

A

um That being said, I, don't I'm, not an expert in topological data analysis, and my impression is- is that there are these methods that exist but they're not like they're, probably new techniques that can be invented and.

D

So that's yeah.

A

C

E

A

On like homology and some of these other geometric approaches.

F

Yeah yeah I couldn't even this my uh my motivation, uh but first of all my motivation is actually I, think uh toological. This analysis have been uh widely applied to the biological stuff, but I think uh one thing: one interesting thing is.

F

They have also been applied to the biological analysis as well, but uh like there is a it's a gap between the TDA and uh depending approaches, especially the proper natural I. Think government. That's one interesting points about graphic natural. Is uh this approach? We consider the data as a set, and that means uh from this kind of perspective you will. You can consider the geometry properties and the topological properties of the data.

F

Yes, so it could be different from the like conversion or natural which consider data from the perspective of visual uh information from the from the images. So this could be different, so so I'm. uh The motivation for me for this project is I want to see if there can be some like interesting stuff happen. If, when we uh like integrates, go to a natural way, the topological analysis, yes on the biological stuff- uh oh maybe I can make myself more clear, like we all know, that's the multicellular system, for example the cell development process.

F

They have Rich topological information and we know that we have a lot of approaches such as the TVA, which are very good ads. uh Exploring the topological information from data like from the microscopic video and we have graphene Network, which are very useful for uh like capture information from the big data from a lot of uh videos when the data is very large data science is very large, so I mean they have uh their own like advantages for the downstream task, but uh actually I'm a graphic.

F

Network guy and I'm more familiar with graphing, Network and I, see that uh in the field of crafting network uh actually I would not say the crafting. Network have been widely applied to the cell development analysis. I will not say that because I didn't see too many examples.

F

So first of all, I would like to see how graphene you know natural can be applied to the cell development analysis and secondly, I would like to see when we consider the topological information in the cell development process, how the graph finance will perform like how will perform with the help, with the supports of topological data analysis and something like this I'm.

F

Looking at some intersection between these two films like and how you would perform on the this multicellular system and from another perspective, I think is I think uh when I said what I want to say is uh I wanna I wanna like explore any kind of uh explanation of the different technique from the geometry and the topological perspective. Yes, like so now, there are some a lot of photos working on the geometry or the topological uh techniques within the context of keep learning and they thought that's.

F

This kind of approaches are very helpful for us to explain why, depending to work- and why does your network would work so I'm also very interested in this. So that is another reason that why I want to look at intersection between the graph and natural and technological analysis, yeah yeah.

A

I was thinking actually, as you were talking about, like a lot of the things that people do in 3D modeling and that's highly. You know uh topological and geometrical. One of the things that you do is you create an object, and then you extrude it in space. So you pull it up. You know you pull it out and you give it some depth.

A

You have to define the depth and what tends to happen is that you can you know, sort of manipulate the shape according to what you want and then so I'm thinking about like some of the things that happen in development, and you know having these sort of Extrusion processes and they're not like you know, they don't call them Extrusion processes in the literatures like Transformations, like imagination or gastrulation, or things where there's a change in shape in the embryo, and so you know it's like they don't describe it.

A

They describe it as like a process and they observe it. But there's like a computational element there. That could be not just Extrusion, but like a lot of these, like cynical complexes that you would sort of discover in a network, you know how do those changes.

A

We change the network, Point Cloud as we get cells, dividing and they're changing their orientation, and so you, basically, if you have like uh cyclical complexes or just like shapes that form in the network, so it could be a triangle or a diamond, and then that diamond would sort of track with the cells dividing and that Network expanding so that you might end up with, like maybe a hexagon that shifts its orientation, maybe even 45 degrees, and then it maybe curves a little bit on the bottom and you keep following that shape out as it's moving around extruding and you know being warped in different ways.

A

So that's you know. That's the kind of thing too that we could think about. Like you know, what is it we start with, maybe like a very simple shape, but a very Sim, maybe an eight cell embryo, maybe like a a diamond of four cells that kind of are in that part of the embryo or maybe like the anterior end of the embryo, and then, as the number of cells increases as you get daughter cells and they change their position, they migrate, they sort of form these structures that start to you know tissues.

A

Those will change the shape and you can actually track that shape. As like an object, that's you know, uh it has properties just like a three-dimensional object in a in something like blender or Unity, or something like that.

A

A

D

F

I'm thinking about.

A

It yeah anyway.

F

A

So that's that's great yeah. Thank you. uh Jihang I think there are a lot of things that like following up from our discussion two weeks ago and I, just yeah I wanted to get the people uh doing. You know I want to get people like camancho and sushma thinking about especially himanshu about, like you know what what? Where are we going with this stuff?

A

So we're going somewhere, uh that's very kind of conceptually hard, and so we want to have one I think a good sort of a good sense of where that is and what's possible. um So.

F

Yeah yeah, yeah, yeah, actually I think uh so again, let's just uh uh because actually for this project, I'm working on graphing network, uh topological data analysis and um less good ads, the like the segmentation or relating stuff, so so, first of all for This research projects. If, like anyone interested in this and I'm, very welcome for everybody uh which reach out to me and another point is uh during this whole jigsaw period. If you have any related questions to the topological analysis and The Graduate Network I'm also welcome for everyone like reach out to.

D

Me yeah: it's pretty awesome.

A

Thank you. Okay, now, I want to explain something that is related to the conversation that Gian hung and I had about networks into topological data analysis in gnns.

A

D

One thing I mentioned was.

E

A

Complex that you find often in Networks, so let's consider a simple Network here, where we have connectivity across five nodes. Okay. So this is a network topology where you have this triangle and these two lines that come out now in a normal network analysis we might be interested in calculating like the centrality or the clustering coefficient or things like that. But in a topological uh network analysis you might be interested in finding shapes in the network, so what kinds of things are enclosed in the network?

A

So in this case we have a triangle which we can Define here and if we add in nodes like.

A

D

A

Another triangle actually two triangles and.

D

E

A

A rectangle which is here, and so these are what they call cyclical complexes.

A

And these are fine. You know people uh are able to discover these, and these are kind of ways of flipping. The typical network analysis on its head and saying, instead of maybe finding communities or finding centrality or finding a connectivity regime, I may be interested in finding shapes that are formed by the network. So this is important in sort of the spatial networks that we're dealing with in the embryo. At this point, Cloud that we talked about earlier, which is in three-dimensional space, they're, actually a four-dimensional space.

A

We have this point cloud of points centroids. These centroids are kind of worked out in this way, so that you have, uh you know they're always changing. We talked about this where this point cloud is always changing: we're getting new uh cell centroids we're removing old cell centroids, and so this topology is always actively evolving.

A

Instead of this point, Cloud actually actively evolved in terms of its outline in terms of its structure and things like that. Okay. So this is the point cloud.

A

So this point cloud is constantly evolving and, as we said in the meeting in the main meeting, um you know we have these two views. We have the static, View and the dynamic View, and so, if we look at the actually the point cloud or anything from those two points of view. Well, let's think about the network.

A

We have a static Network, which is again something like this, where we have a triangle. This is our triangle and say that triangle changes over time. So maybe this node kind of comes out here and gives birth to two daughter nodes.

A

So we have something like this so This node here is here: it's now removed and you have two daughter nodes in its place. That means the connectivity changes so instead of a triangle, you end up with this sort of rectangle with two means here, and maybe this node actually dies and you end up with daughter cells here, where you end up with maybe a not a triangle, but like a fan motif, let's get this fan and then this might be connected later into two more triangles.

A

But let's focus on this this image. This part, for now you have this triangle and it transforms into a rectangle, so that transformation then is dynamic.

A

So we can take two static views of the network on a different two of different points in time and end up with this Dynamic shift from a triangle to a rectangle or rhomboid I guess would be a better way to describe it, and so these two shapes this transformation of shape is actually what we're dealing with in development, because we have our embryo and it's changing its shape uh constantly, as those cells are dividing, so we've removed from the point Cloud to these kind of structures.

A

Now one thing I was talking about was like say: we have like a truck or a like. A diamond is a cyclical structure. You know we have this four, this four topical symbolical structure, I, guess you define it as you have four nodes here and this this diamond might not change shape to something else.

A

It might actually transform its sort of form. So like this thing here is flat. This isn't two dimensions. Let's imagine that the cells change your position.

A

You get daughter cells that emerge around it, so there's a migration of cells. There's a change in the constituents of the network. You.

E

Have a diamond: that's worked into a three-dimensional.

A

Shape that now has sort of it's extruded out in this uh in this plane and then also in this plane. So it's kind of hard to see. I've also made a mistake in my drawing, but basically you have this shape. That's changing its form, and so this is where, where I was kind of going with this, this is in 2D, and this is a static View all right, and then this is where you're going into three dimensions. Where there's Extrusion.

A

And that Extrusion maps to some developmental transformation, like maybe something like imagination,.

A

Or some process um like gastrulation.

A

And so we can actually take these geometric Transformations map, the developmental Transformations, and this actually goes back to Darcy Thompson's work way back in 1917, um and you know, the sort of the idea was that you had like you could take a fish and you could put it on a grid where you could. Each Landmark would have like a certain point on the grid, and that tie point then, would would allow you to transform the Grid or transform the phenotype and you'd have a corresponding mathematical structure or you'd have a corresponding morphological structure.

A

So it was like a morphological structure like the fish, so I'll just draw this out. You have a fish like this. This is a very.

D

Simple problem, like some sort of fish.

A

The point here is to show like with this phenotype phenotypic transformation looks like so we have rays which are kind of overlapping here, and you have this phenotype and you have this grid behind it. So this phenotype is embedded in a grid.

A

This grid has tie points.

A

Basically, where the landmarks are so the script goes underneath the morphology and there's a lot. You know this is a pretty simple representation, and so you might have type minus the eye. You might have tied points like this under winding the fish, and so each of these tie points corresponds to a certain Landmark on the fish's morphology is.

E

A

Morphology changes, if you look at say like, for example, another species. These changes move around this grid, so the grid gets warped, so the grid actually changes its form. So it goes from something straight like this.

A

To something like this.

A

And so you can see what the transfer topological transformation looks like it's.

E

A

From a linear to a non-linear sort of way- and this is of course uh changes in species- but it could be changes in time- it could be a develop, it could be a developmental phenotype like an egg or a some sort of in a learnable stage uh organism.

A

So we can do this sort of uh warping with developmental forms, and then we can look at this relationship between uh embryo networks or some sort of network topology and these shapes and how they change their form over time.

A

E

Yeah we have any.

A

Comments about that or questions I, don't know if how dick is doing, uh how are you doing dick.

D

I'm doing better today, good, that's good yeah! uh If you want they'll uh present that uh Strange New Image, processing problem, okay, yeah that'll be good. Actually, okay, uh I came up and came across it uh uh because there are now uh uh some x-ray sources which will produce a narrow beam of light or R of x-rays.

D

Okay and uh I realized that in trying to solve the problem, there's a much simpler problem which apparently has not been addressed in image processing. So uh let me let me uh share my screen.

D

Whoops make sure my screen actions.

D

Okay, uh entire screen and back up here: okay, yes, yeah, okay, okay, now I'll switch to Auto draw to show you what the problem is start crying. Okay, take any picture whatsoever which I'll represent as a square and let's draw a line across it. Okay, suppose that we sample all the pixels along the line.

D

Okay, then go another way and do the same thing. We sample all the pixels along the line and another one and another one okay. Now what this does? Is it Glides the pictures into regions?

D

And we could start to do in what's called in in painting or in filling okay interpolating between the values on the lines this one that produce a picture of some quality, uh and if we have enough lines, I mean it's clear here. Let me do another one here, it's clear from the old-fashioned CRT uh televisions, which, if you have enough lines, you can get a good representation of a picture. That's what television was based on originally until the flat screens have shown up right, uh and so those are all parallel lines.

D

But I'm, saying let's take a random set of lines. It was an in-painting algorithm and that produces a picture, and the question, then, is: uh what is the quality of the image versus the number of lines?

D

Okay, now the lines can be random uh or you can do it more cleverly and say we're going to add a line, but it depends on the quality and the image and where we put it, so you could then have image dependent addition and lines which will then do it within painting and then you'll be further Improvement in the picture.

D

So the question is: what random lines are linguistic image dependent lines? What is the quality of the image versus the number one?

D

Okay, that's that's the general question.

D

Okay and it's it's a it's a it's a bit strange because you're taking these lines- and you know, function along each line which the value of the pixels and uh and then here's the infilling algorithm to fill in the picture.

D

Well, this there was an old, an old problem of compressing images for transmission, so they went much faster. It's not a problem nowadays, uh but this would also your particular solution to any progression and I can't find anything in the literature except you know, International TV screens, uh which do anything like this.

D

So it seems to be a new branch of image processes.

D

Okay, so uh I'm looking for something to work with me on testing this idea out and see where it goes. Okay,.

A

D

This is abstract it, but it came out of the X-ray sources which generate lines across objects if these will be x-ray lines instead of the visible lifelines, but backing up. It's obvious that the general problem, this combination of lines Plus in painting, is a form of representing an image with a lot less data than the whole image.

D

Okay, so that's the problem.

A

C

A

We've talked about that with respect to the breast cancer work and how.

D

Yeah yeah the cancer work inspired, but I'm sort of backing up and saying, there's a general problem here and if we can solve the general problem, we won't let the X-ray problem also.

A

D

Yeah, so this is a more General sampling.

A

Problem almost.

D

Yeah, it's more General because you get all the X's along the line, yeah because x-rays, you only get the sum of them.

D

A

D

Yeah I'm, seeing if I can jog Jesse here, oh yeah,.

D

I'm looking for a partner in this uh who wants to it's much simpler image processing than these segmentation problems? Yeah, okay, but it seems to be a brand new uh topic for image processing.

D

And who knows where it'll go.

A

For one thing, yeah one area I could imagine immediately is imagine if you had a an image of a lot of noise in it and you couldn't uh unmasks yeah. So you had things like a bunch of noise as a mask over an image, and you couldn't tell what the signal and noise like you couldn't decouple them. So you know you'd have to do that. You could do this kind of sampling where you go over basically you'd end up sampling things at random.

A

You just kind of pick a pixel and you okay, flip it over and say: what's the content, information content of this pixel or that pixel you did it this way, you could like, uh like kind of divide up the image, so you could send a signal across. You could send a signal across here and.

C

A

Could find the intersection after you have enough lines and like explore that space? It's almost like an.

D

Example of an application of this would be if you use multiple satellites and send signals between them and the signals intersect atmosphere. You can reconstruct the atom and from the seven lines between satellites yeah. That's an example of a practical application.

D

Another one is geological: if you have a shaft and you you drop a and you slowly drop a transmitter of some sort, and then you have another shaft at some distance from it and you raise and lower a receiver.

D

Then you get lines between the two shafts and you can reconstruct the Earth between the two Jets from the set of lines: yeah. Okay, that's another example: okay, so uh there might be some a couple of practical applications: yeah, that's just the old comparison, Village compression right.

A

D

Almost like a dividing Contour.

A

Algorithm on a space like that.

D

Yeah I mean image compression we solved by just increasing the bandwidth.

D

You don't bother with it, okay, but uh it's still curious, It's, a curious problem, reconstructing an image from Lion's glove.

A

A

uh People: okay, oh yeah, he dropped it off there a little bit.

A

Did you did you? What is the last thing you said you dropped off.

A

Okay, all right! Well, thank you! That's great, okay, sure, all right. uh Anyone else wanna have any updates or Susan or Jesse.

A

B

um My my Imaging problem is a bit of a strange one: um I have um X Y and Z axis and I have let's see this is this is the z-axis and this is the x-axis and then I have the y-axis going this way here. I, don't know if you can see that yeah yeah and then um so. My diagrams are 90 degrees flipped from the paper I'm trying to follow and I suspect that they're they're y-axis is going out of the page instead of into the page, and so things are 90 degree flipped.

B

Does that make any sense to everybody here? Does that would that flip and graph of.

A

I, don't know, would it flip a graph or I, don't know how yeah.

A

Anyways yeah I, don't know uh what what program are you using.

D

A

D

A

Just plotting in that way it's a little bit different than like, so like. Usually, the Z is the depth, but this is you're.

B

Saying yeah, but if you, if you have this, where you have so your X, this is the C, and this is the Y you can flip it so that that that's the Z is coming out of the page. The Y is up and the x-axis is is going um that way. So it's a more conventional system right right, I mean it's rotated.

A

Wow yeah yeah I, don't yeah I, don't know I've not run into that. I guess and Manchu said it would flip a graph I think, but both whatever let's see yeah, it would flip a graph I think, but both uh Y and Z would be affected right.

A

B

So um Amy could I show you show you.

A

Yeah yeah, you can.

B

A

Is that better, you can see yeah, you can see your desktop now, so it should be. If you see mine there, we go yes, okay,.

B

Okay, yeah now so I think this graph that looks like a mountain here. A pink mountain um I was trying to duplicate this and I get graphs that are like this.

B

On the left, hand, side and I took this blue line here this uh it's at 0.4 stream and I, moved it around the axis and just changed the negative of positive and I. Still I don't get I. It's got a curve going up, doesn't doesn't have this curve going down like this; in other words, it's kind of 90 degrees flecked. It needs to be turned over right um and then I I looked at this, and this is rubber and it looks kind of like, like my curve and the pot, the first quadrant um anyways.

B

um You know trying to find the other one. Oh and then I have this, and this is for a tall prism instead of a short one. Anyway. I got this to work for me um here, because I had my numbering system backwards.

B

A

B

Yeah yeah, it's interesting.

B

Memory system is bad guys here here, I, don't know if you can see it in the corners, the um X, X, Y and Z, and the Y is going into the plane. Okay,.

A

B

And if you somehow number it so that the Y is going out of the plane, then you seem to get um a flipped curve.

A

Oh so the the number, the number.

C

A

uh It's basically flipped from positive to negative, or it's negative, going versus positive, going yeah.

B

A

B

Anyways, that's, it should be holding me up. Couldn't.

A

B

Many weeks working on this and I keep getting this type of thing. Yeah.

B

It needs 90 degrees slipping anyway. It's a really stupid problem, but uh it's consistent I get it this this one for a really tall um triangle prism. This is sort of a medium tall one, and this is for a shorter one. Yeah. You can see. There's a theme here: okay, not it doesn't follow this pill. It's not a hill, it's more of a swirled right right and that's my problem, so I, just it's stupid and I'm hung up because I'm supposed to duplicate this paper or explain why it doesn't work.

B

So any thoughts on this would be would be really great. Yeah.

A

Yeah I think that's some interesting data and it's always hard to like get a uh to get the graphs to plot correctly because they use their own conventions. Like um you know, each.

C

Program uses kind.

A

Of a weighted that it puts in data so, like you know, your typical, like Matlab approach, is to use like a vector, or you know, some structure is some struct that you can plug in the values to the axes.

A

For the XYZ, though, uh couldn't you just uh transform the y-axis to something it's like.

B

Yeah I was thinking of just turning it there's a method of doing it, so I'm just gonna.

A

B

um Change that around, if I, can figure out the code like there's hidden, um there's a whole pile of hidden attributes in com console, um it tries to do everything and, from my perspective, it ends up doing nothing.

A

E

I thinking just.

A

A simpler case of like one over X, and that would give you like a like a decimal, but it would be like transformed y, so it would like turn it around, but it would be, and it would be a weird number uh like the weird interval, but it would be transformed y, so it'd basically give you that orientation and you just well. This is transformed. Why and it's going in a different direction. So I don't know if that would work, but something.

B

Like that, I think that that console does have it where the y y direction can be changed. Okay, um yeah I was like I. Looked it up and there are, you can even put it in a spherical coordinate system and if you get desperate, maybe I should just do that.

A

Maybe yeah it's just gonna no.

B

Yeah, you know.

A

It's more General right now, when you rotate them and I, don't know if you're going to be rotating them in the presentation and what but that's another thing to consider. Yeah.

B

um That's why I've been having problems? Yeah, yeah, that's exactly why I've been having problems.

B

uh I'll stop sharing okay,.

A

Well, good! Luck with that problem! I, don't know, I, don't know what else to say about it. Yeah! Those are some nice graphs, though that was a good good uh collection.

B

Yeah at least they're consistent yeah, going in the wrong direction. Yeah.

A

It's all right, uh I had a few things to share her myself. um So let me share my screen or at the top of the hour, but it shouldn't take that long to talk about um so one thing I found this week was this interesting. This is having to do with uh using genomic data as sort of like an input for what they call Foundation models.

A

So this is I, think from Stanford uh research group at Stanford and what they're doing is they're doing they're working with hyena DNA.

A

um What they're trying to do is build what they call long-range genomic Foundation model. So this is a model where you can train a model on DNA uh like genomic.

E

A

A bunch of nucleotides and it can predict or make predictions for different types of uh DNA stretches. So one of the problems that we've had in in genomics is DNA alignments, uh dino, weinmans and then discovering stretches of DNA. So we've in the past, we've used things like blast. We use in all sorts of mapping software where we're mapping between a reference genome and a genome that we're sequencing and to get those aligned to get them uh to know where they are in the in the reference. So this is a a little post on this.

A

This is we're excited to introduce hyena DNA, a long-range genomic Foundation model, with context links up to 1 million tokens that single nucleotide resolution. So this is just a term. That's you know. Tokens are like what these in this model to um you know, provide this context thing um hyper hyena DNA is pre-trained on the human reference genome and sets new SAT a on 23 Downstream tasks.

A

This is a I think state of the art, um including predicting regulatory elements, chromatin profiles in species classification, so they can take a like. They have a reference genome that they train the model.

A

On and then it's it's kind of like a large language model where you're predicting on a bunch of letters, you're finding uh features like regulatory sequences, chromatin profiles, if you want to find a chromatin uh different sites for chromatin activity and species classification, so you can detect like a certain stretch of DNA, and usually we do this in Blast, where we find potential matches for some piece of DNA that we put into a blast search.

A

So this is you know this will give you those predictions I, don't know how it compares the blast, they don't really say, but basically that's what they're trying to do. It's like kind of a Next, Generation blast and kind of using a large language model to or a large language model techniques to sort of get at this problem.

A

uh We also explore what new capabilities open up with long context and genomics, including in-context learning with soft, prompt, tunable tokens and instruction fine-tuning so.

E

This is the archive.

A

Paper here uh this is hyena DNA. They talk about the paper. uh This is what they call it a foundation model.

A

So, uh let's see they do actually have some interesting things with respect to other uh types of machine learning models, um so leveraging hyenas new long-range capabilities represent hyena DNA. They talk about hyenas.

E

A

Here, which is an approach that they've developed, hyena large language model based on implicit convolutions, was shown to match attention, which is a method that they use in machine learning and deep learning in quality, while allowing longer context lengths and lower time complexity. So it's a large language model that they're using they're fine-tuning it with respect to the data here, leveraging Hyena's new long-range capabilities, we present hyena DNA. So this is a subset of hyena hyenas, which is the model that they have here.

A

A genomic Foundation model pre-trained on the human reference genome with context links up to 1 million tokens at the single nucleotide level. So they train at a single nucleotides and up to 500 500 fold increase over previous density tension-based models. So people were using attention-based models for some of these genomic problems earlier in the so, people have been applying machine learning and deep learning to this uh to sort of I guess: sort of get improvements over black traditional blast, searches and they're, actually improving over these attention-based models.

A

uh Hyena DNA scale, Sub quadratically in sequence, length, training up to 160 times faster than Transformer use a single nucleotide tokens and has full Global context in each layer. We explore what longer context enables so what that means is. It includes first use of income tax learning in genomics for a simple adaptation and novel tasks without updating retrained model needs, so basically they can find uh you know very specific sequences, they can find novel sequences and it doesn't need a lot of updating in terms of training.

A

So this is a pretty flexible model on fine-tune benchmarks from the nucleotide Transformer ion and DNA reaches state of the art, which is so a on 12 of the 17 data sets used using a model with order as a magnitude plus parameters and pre-training data. So.

E

They're able to.

A

Train it on you, know their reference genome they're, able to get a good result on 12 of 17 data sets that they tested against and they're able to use fewer parameters and pre-training data in general. So this is an improvement on some of these methods. So this is really about.

A

You know getting this um kind of get going from like blast searches which were kind of so you know, 2000s to uh you, know different machine learning techniques to now this type of model, so um they were talking about long context, models which are uh state-of-the-art thing. It's very uh trendy opening I stated a goal to reach 1 million tokens. These are these long context models, while magic announced they've reached 5 million tokens for code world for it, uh but well most of the focused on natural language and code.

A

So this is where a lot of large language models focus on like using English and predicting text in English they're, using something in biology, they're using DNA sequences, um and so we have of course, four nucleotides that get rearranged from different ways, and you have these long stretches of DNA and there are actually quite a few possible combinations that it could be. So it's a hard problem to solve and you have, like you know things within that those stretches of DNA, so you have different patterns.

A

uh You know you have repeats, you have of different lengths. You have different types of uh combinations of codons and things like that. You have stopped codons, you have Stark codons, and so you have a lot of things in the structure of DNA sequences, which need to be predicted.

E

And so it is like kind.

A

Of like people have made the analogy between DNA and and language before, but this is where you're actually using this to find uh to to improve the search process so uh to give a sense of the scale. The human genome is 3.2 billion nucleotides uh and can be seen as characters in the sequence so they're able to.

A

They have some blog posts on this as well, um so they they train this hyena model, which is a convolutional large language model, um and so they've actually been able to do this with, or you know more generally, but they've applied it now to genomics where they want to find. uh You know they want to really kind of put this to the test, so here's the sample from the so they have a reference Gene. You know, usually you do this some sort of reference genome that you build so the first team in genome.

A

uh You know: first, human draft of the human genome is kind of like a reference genome and they do this for different species where they kind of sequence, a bunch of organisms, and they they find a consensus for the sequence. And then they they put a draft out.

C

And then that's.

A

The the uh sort of the reference genome and then, if you just go to sample a genome from your own sample, that's your sample and then you try to map it to the reference genome, so you can actually align things and make sure that they're in the right place in the genome. So that's what they're trying to do, but they're trying to pre-train the model with a reference genome and then throw data at it from different samples and see if they can make the predictions. So this is how it works. Basically, they have this reference.

A

Gene now I'm going to use that as the pre-training step. Then they have these predictions, these token predictions that they they propose. They match it up with this technique here uh and then they have these Downstream tasks where they try to classify things structures within the DNA sequences like finding regulatory Elements, which are these you know sometimes they're repeats, sometimes they're, very specific sequences um domains in within the DNA.

A

Then they have chromatin profiles. So you know you might have a certain two or three DNA sequence or nucleotide sequence, Motif within your DNA sequence, that you can discover, and these are potential sites for chromatin binding. Then you have in context learning which are you know, other features that you want to learn about, and that's really, you know sort of the the heart of genomics.

A

That's why we we do genomics is because we want to find these patterns that we want to say like this is useful, for you know some biological function or you know it's, maybe a new gene, or something like that. So you know this is kind of they're, really kind of moving this area forward um and so they're trying to learn all sorts of things uh they want to learn, maybe to Define, like coding, regions from non-coding regions. They want to find out more about gene expression.

A

They want to find out about other types of structures in the DNA and they ultimately want to go to building. uh You know going from DNA to proteins. So you know this is where you have a sequence of nucleotides that gets transcribed into RNA, which has you know usually there's some changes in the RNA that uh derive from that transcriptional process than that RNA gets built into proteins through another process of translation, and so you know this is the first step in predicting what these proteins will look like.

A

If you can predict patterns in the DNA, you can then take samples of RNA predict patterns and you can use something like rna-seq or something like that predict patterns in the RNA and then link those two things and then link that to the proteins. Because now you know we can of course model in sequence, proteins. We can actually make predictions on that and I think people are applying uh larger language models to proteins as well I'm, not familiar with how that's you know with the state of that field is.

A

But um so you know you have these different DNA motifs promoter silencers enhancers, and you want to be able to discover those in the sequencing context.

A

So I'll provide this paper uh where this is in the slack and if I can send it out an email as well. People want to read it.

A

um Let's see. uh So let me check the uh chat here. Okay, so Jia hung had to leave uh and if you're interested in what jiahang had to say about networks and topology, don't hesitate to reach out. He put his email in the chat um if you're interested in TDA and gnns and yeah.

E

A

Keep us updated on the graphs.

A

So I think that's it for today any other comments or questions.

A

I'm, good, okay, all right! Well, thanks everyone for your uh discussions. It is very good, very good meeting.

A

Have a good week, thank you. Bye! All right now, I'd like to talk about uh sort of a third installment in the discussion we've been having about different types of embryoids.

A

So two weeks ago we talked about human embryoids and some of the work that's been happening there and then last week we talked about monkey embryoids and non-human primates, so we have monkeys and humans, and so we've been able to demonstrate this in two model: organisms in in primates uh this week, I'm going to talk more about human embryoids and some of the gastroloid work.

A

That's been done so you know this is like I I showed a diagram two weeks ago, showing sort of this progression from organoids to gastroides, to embryoids and just kind of how those models are. You know they're kind of refining those models to give us answers to different things. We could look at different things in development and get certain answers depending on the type of model. We build. There's a lot of control here. It's either transgene control, transgenic control, which means you put a gene into the cells. You maybe transform them into an induced portal.

A

Potent stake, or you put a transgene in that you can induce and you can make these changes.

A

You can also do things with defined media which we'll talk about today, so the Hana paper from the first week that talked about defined media and kind of getting the perfect formula for your medium and then using that. Then this is something that goes back to and do support you know making into supporting potent cells where the some people use transgenes. Some people use a defined medium of a certain type and I can tell you that there's a lot of sort of technical uh gamesmanship there.

A

So you know finding the right medium is great, but some people like there's a lot of skepticism surrounding some of the results for both transgenes and and define media. So you know when you read these results, it's amazing stuff, but you can also ask the question you know: are we really getting the thing? We think we are we're getting you know, differentiated cells, we're getting tissues, we're getting things like that.

E

But is this really translatable.

A

To say, like a whole embryo setting, where you'd actually be able to make uh some priming or maybe clone a primate- or uh you know, creative primate embryo that could be implanted, which we talked about in the first week as well.

A

So with that in mind, let's go to the papers, so this is actually an article in I. Think it's in the Observer It's a UK publication. This is by Philip ball. This is a feature that they had on this. This human embryo models were so this is kind of considering some of what's been going on. They showed an example of how these embryo models are made. So these are embryo models, they're, not necessarily embryos, they're, technically embryoids or they're, technically something to some of the embryoids.

A

They could be a gastroyed or uh even just the standard organoid. Although what they're trying to do is build this embryo with the tissues and with the structure of an embryo as opposed to just creating, uh maybe like Joe Myers, but they are creating germ layers as well as we talked about in the first and second weeks. So you know they're creating these embryo models out of stem cells, they're transforming them to these different Fates building.

A

You know there is a building, differentiated tissues, they're, building, a structure that looks like an embryo and so that that's where we are and they're a host of ethical challenges surrounding this I, don't want to Discount those. This article I just talked about these um rare to see like a print version of an article anymore, and you can't really read the text but basically goes over the papers we talked about in the first week and and some of the so you can find that article online. You know that would be weird about it.

A

uh So the first paper I want to talk about is this paper. um This is uh a bunch of authors here, um just a lot that all and the title was modeling human post-implantation development by extra embryonic Nation engineering. So this is the work on defined media and the idea behind defined media is that you have the cells grow on this medium and you have to find the right recipe. So it's a chemical inducement of this to create these underneath holes.

A

uh You've had a friendly right chemical formula, and so there are all sorts of growth factors in media there's.

A

All sorts of uh there are all sorts of chemicals that keep the pH uh at the right level, and things like that, so they're all sorts of things that we can modify to make optimize that environment and that's what they're doing in this paper um they're trying to kind of they're talking about this in terms of extra embryonic Niche engineering, so extra embryonic just means out in the environment of the cell or of the embryo, and so when you're out in the environment,.

D

A

In a biological.

A

Or something like that now, the environment is, you know we talked about environment before in the group and that is to say that it's hard to Define, you have to really Define what you mean by environment, which is of course a concern here, because this is a chemicals formula for the environment. Basically, so this is. Where uh kind of you know, this is one approach to making these.

A

So the abstract reads: implantation of the human embryo commences a critical developmental stage that comprises profound morphogenetic, alteration of embryonic and extreme embryonic tissues, access formation and gastric relation events, so you're not only creating the tissues you're, creating the axial polarity, so things are towards the header towards the tail. In that sense, um you know so they have to be in the right order.

A

They have to be in the right place, not just differentiated many of these gastrulation events, so, like I, talked about in the case of some of our topological networks, you have these shape, Transformations that you know need to happen and a very specific shape transformations.

A

Our mechanistic knowledge of this window of human life remains limited due to restricted access to. In Vivo samples, well Technical and ethical reasons, so we can't necessarily create, like uh you know something in the womb uh that's frowned upon, but we can create it in a three-dimensional culture. This.

D

Is why we were interested.

A

In every models, because we can't experiment on fetuses basically- and uh you know- there's a lot of you know that's. This is actually why we have abuse policy cells, because there were a lot of ethical concerns and people have moral objections. Sometimes the stem cell models actually harvesting them from the embryo, and so that's where a lot of stem.

D

Cells come from.

A

They come from umbilical cord, they come from the embryo, and so this.

E

A

Thing, that's been, uh you know at the floor of this field for a long time, so uh this gives us a model where we can work on these problems uh without some of these ethical concerns.

A

Additionally, human stem cell models of early post-implantation development with those embryonic and extra embryonic tissue morphogenesis are lacking. So we don't really know anything like I said in the first week about posting Plantation development. We know a lot about pre-implantation development in terms of like working on it and culture and models, but this posting Plantation world is quite different here.

E

We present I discoid.

A

So it's a small eye discoid and it sounds like something Apple would make, but it's actually a biological thing uh here we present I discoid produced from human inducible and stem cells by an engineered synthetic Gene circuit, so they're actually using uh uh a synthetic Gene circuit to sort of helpless. uh You know help the embryoid model uh uh respond to its environment, the optimal environmental conditions. So you have this uh you're, basically creating an inducibility stem cell you're using an engineered, a synthetic Gene circuit you're using that to uh sort of control.

A

The fate of the cell um I discoides, exhibit reciprocal co-development of human embryonic tissue, an engineered extreme, organic niche. Remember all these chemicals in a model of human post-implantation, the end they exhibit unanticipated, self-organization and tissue boundary formation that recapitulates the oak Sac like tissue specification.

A

So this is at a certain point in development where we get this yolk Sac, we get the tissue specification, the boundaries of tissue start to form, and this is all sort of self-organized. There's. No, you know they're chemical signals within the embryo, but particularly their chemical signals outside the embryo that are sort of enforcing this.

A

So these yolks act, like tissue specifications, are extra embryonic mesoderm in the hemoplietic characteristics, metapoetic characteristics being the formation of blood I think we talked about that in the first week. That was one of their Hallmarks is where they were actually developing.

A

uh They were looking at how the blood the the circulatory system develops in the formation of blood. So hematopoetic characteristics are the formation of bilamina dislike embryonic morphology, which is you know your uh red and white blood cells. I. Think the red blood cells are what they focused on in the papers. In the first week, the development of the amniotic like cavity and acquisition of an anterior like hypoglass pole and posterior, like axis idisk, leads offered easy to use high throughput reproducible and scalable platform to promote multifaceted aspects of human early posting implantation development.

A

So we can do a lot of things with this. We can do drug testing, developmental toxicology and modeling diseases that occur in utero. So here are some images of what they're doing here. This is where they have zero percent I get a six highest supplementation. This is okay, so this shows some of the features in here. Here's the hemoglobin starting to form here. This is uh another example of hemoglobin.

A

This is day 12 of development, so we talked about this being the window on these embryoids are probably it you know going to give you the best window in the development.

F

So you have some hemoglobin.

A

Signal here uh you have some again six High supplementation 25, so it's a zero percent. This is 25 percent. This is showing some of these hemoglobin in context here. So here's some of the structures here in the hemoglobin surrounding it.

A

uh Here's some more pictures of where you have kind of a you're going through the structure from the mesoderm to the endothelial, to the endoderm, so from the bottom to the top and here's some images here where you have stained for Fox A2, cd34 Desmond, which is a muscle thing, an effect in which is another muscle thing. So you see a Desmond is in here being expressed. uh Fox A2 is in the blue, see some of this in the endothelial, and especially in the endoderm and.

E

The mesoderm means.

A

You have a lot of Desmond, which decreases as you go towards the endoderm and then cd34, which is another marker, which decreases as you go towards the endoderber, the top. So you get these three uh germ layers, these three types of precursors to the tissues and you get different things expressed.

A

So you get things expressed with respect to muscle and some other things that are in so you can see that they differ where you are in the sculpture same thing here you have Desmond and white, you have cd43 and yellow and then cd31 and red, and you see that as you move through uh from the mesoderm to end of the oil hematopoietic cells, that it's changing, it's uh composition.

A

So this is just an example of how these things form in time. This is actually an example here of what's happening in these embryoid models. This is a figure where they show the first symmetry breaking here, and then you see confinement and migration, so cell migration is happening, and then you have this bilometer boundary, so this is uh corresponding to differential gene expression, so you have so. First of all, you start off with these two cell lines. The I got a six which are uninduced in the wild type.

A

Then in the I-discoid cell line you have the combination of them, so you're inducing the I get a six, but it's also mixed with mild type cells. You have the Symmetry breaking between the I get a six and the wild type. We have this confinement of migration of different I get a six cells. You have this bilaminar boundary between the Gata 6 being expressed here and then the wild type you get nanog expression, which is these are.

E

A

Of course, an important marker of differentiation uh and stemness.

E

A

Is where you get more, uh this sort.

E

Of lumen forming.

A

In blue, this is another marker that they used for that and the unlike formation. You have the bilaminer disc and you get another structure here, the Amion amnion.

A

So this is like the amniotic uh SAC and the disc associated with it um the posterior formation. Here, then, this is the model of the human yolk sac and you're modeling hematopoiesis here uh from the music. So these are all things where you're starting to get structure you're starting to get different structures. You recognize as embryonic especially human embryonic, and it leads to this model.

A

So the second paper I want to talk about is self-patterning of human stem cells into post-implantation lineages.

A

So this is from nature, and this is an accelerated article preview, so they're going to have a different version of this out soon, where it's the final version. This is weird the way they do this in this part of science, where it's very competitive. They have these a borrowed papers. They have accelerated papers, they try to bring them on all at the same time. It's just. This is part of the culture.

A

um This article, uh we'll just go through the abstract, so the abstract reads: investigating human development is a significant scientific challenge, either the technical and ethical limitations of working with embryonic samples.

A

um So we've talked about how these embryoids provide an alternative to accessing experimental models that we couldn't exist excess before um here we show the human stem cells can be triggered to self-organize into three-dimensional structures that recapitulate some key spatio temporal events of the human proposed implantation of organic development. So this is where we're focusing in these spatio temporal events, this organization and spaces you saw in the previous paper and then over time.

A

Importantly, our system reproducibility captures spontaneous differentiation in code development of embryonic epoblast, an extra organic hypovlastic lineages. So we have two different lineages that are.

D

Differentiating.

A

Spontaneously in co-developing, this establishes key signaling hubs of secreted modulators in an undergoes symmetry breaking like events, so symmetry braking is where you have things that are very similar and.

D

They're identical I.

A

Guess in terms of their spatial distribution, and then you get the Symmetry breaking, which is where they differ, so, like a left and right, symmetry braking would be where the left side differs from the right side, there's an event that triggers a difference, uh and these are common in development. This is involving. You know these signaling hubs, where you get secretions of different chemicals out of the environment, into the embryonic environment. So.

E

A

The medium, the extreme moronic environment, this is within the embryo, the environment, and you get these secreted modulators that get uh pumped out into the uh local environment and it tells other cells kind of what to do so. You want to establish these key signaling hubs. It helps to sort of organize the model, so self-organization, maybe is a little fuzzy in this context, because self-organization just means that it's not involving some sort of supervised process, there's no top-down process, it's just kind of that. These signaling UPS start to form you get the Supreme.

A

This is a secret tone that sort of exists in the model and then the secretion, those secreted molecules actually help to you- know signal other cells and tell them where to go. What to do so. These single scale cell transcriptomics to confirm differentiation in diverse cell States, so we were using a lot of labeling in the last paper, a lot of fluorescent labeling in this case they're using single cell transcript. So it makes this confirms a lot of these sort of sulfates that they're actually forming different types of cells.

A

Now how much of that type of cell? It is is something we don't know, so we don't know if it's exactly like the cells you see in the embryo or something that resembles them um so.

E

A

An important point, because sometimes these cells aren't identical to what we find in an embryo and as a result, we may not may or may not be able to use them for certain end goals like drug Discovery or whatever. So this is uh these papers. I have very detailed figures, and so this.

D

Is just a sample.

A

Of what they have to offer here in terms of their evidence, uh you have a couple things. First of all, I want to highlight the epiblast and hypoblast layers that that form first and then you end up with these more detailed structures. So here's an example here, A Carnegie stage for you of this implantation stage where you have the Inner Cell Mass in CS5, you have this epiblast and hyperblast form around within a trophoblast, and then you have the amnion forming and you have this more detailed structure.

A

So you can see that there's this sort of layering and pattern of formation that accompanies this protocol, um then they show some stains here, go to the second figure, where they actually do some Gene Expressions. So they're doing this gene expression, profiling and they're, plotting it out in a umat space here, so they're, actually showing some of these cell types in a three-dimensional three components of a umap analysis, uh so they're, showing that their differences in the expression profiles that there's some overlap.

A

But it largely is, you know, clustered in the space, so we'll call it clustering per se and emap. But this shows another analysis on two umap components where they show some of these cell morphologies, and this is actually shows like some of these differences where you have mesoderm here. You have extra embryonic mesoderm here the blood progenitors, as we saw in the last paper trophoblasts.

A

Epiblasts and hyperblasts, so these are all separated out in this umap space. So this is dimensionality reduction, you're, basically trying to find how they you know if they're separate or if they're sort of commingled with each other and that's how you can tell generally how you know different the different label cell types are, so they may identify the cell types from some other Criterion like a staining for markers, and then they look at this gene expression profile, which includes email, thousands of transcripts, and they can actually do this sort of analysis. So.

E

A

Good that we have uh the next gen sequencing and the dimensionality reduction here, um and then this just shows more examples of some of the structures that are forming in the embryoid. You have these different states for effect and gappy and AFP. So you can look at like how some of these things are expressed in the structure and.

A

So this is an example here in G of where you get this sort of process here, so you get the hpscs, you get the self-assembly stage, which is where you know they kind of come together in different ways. This self-sorting stage so they're kind of forming an aggregate here from these individual cells.

A

Then they're self-sorting into these layers, they're self-patterning, where the layers start to pattern, and then you get this these secret, these uh secretion centers here and here and then these secreted inducers are pumping Out, secret, home and they're inducing other cells to form other things. That's basically the process here. So you can see as you get this aggregation you get patterning, then you get further patterning from sort of these hubs of secretum, and so this shows again what this looks like.

A

You have EPA blast, like cells hypoglass, like cells, they get transformed through fgf and middle actions in this sort of aggregation step here, and then you get the Sorting of cells. So once you get an aggregation of these cells, you know they're sort of from Mosaic. They get sorted into these layers by uh gene expression profile or by how much fgf and nodal they're producing, and then they start to form this patterning, which needs to be Senators for me.

A

So the hypoglass like cells, either go to this anterior visceral endoderm, like cell fate, or this amnion like fate and then there's a it goes to a mesoderm like thing. If it's given enough bmp4, uh there's feedback bmp4 for amnion, like cells and then emp2 from hyperblast amnion like so um and so they're, you know, bmp2 is playing a big role here in some of these transitions.

A

E

A

Thing to note about this is that they use the term like so epiblescent Lake amnion like ezaderm, like those are all I, guess, they're hedge terms, but they're also very predict they're, also very descriptive in the sense that they're like those cells, but they aren't necessarily those cells we're just assuming that they are based on their gene expression profiles and the way they behave.

A

So you can see this human embryo model here, where you see this uh sort of diagram. This process diagram translated into the topology of an early embryo.

A

And then the final paper is this paper on gastroloids: this is in stem cell reviews and reports. This is on these gastroloid models, so I said that they're embryoids and then their gastroloids, which are a little bit more specific. So these are novel systems for disease, modeling and drug testing. So.

E

A

Just going to go over the abstract here, this is a review paper, so I'll make this available to people in the group. uh Just kind of goes over. The state of the art in this field of gastroint.

A

So by virtue of inaccessible nature, mammalian implantation stage development, which is what we've been talking about, has remain one of the most enigmatic and hard to investigate periods of riogenesis derived from pluribose stem cells. Gastroise recapitulate key aspects of gastro with Stage embryos and have emerged as a powerful in vitro tool to study the architectural features of early post-implantation and versions.

A

So they're, focusing on the stage with these gastroides- and you know it's just really kind of recapitulating this process, which is actually very important to understand, and you can see that there's a lot of self-organization and there's a lot of things going on. So it's important to have a very specific biological model that we can use for this.

A

While the majority of the work in the emerging field is focused on the use of gastro lights, to model, embryogenesis retractable, nature and suitability for high throughput, scaling is presented in unprecedented opportunity to investigate both Developmental and environmental aberrations to the embryo as they occur in vitro. So we want to look at some of these aberrations and development. Maybe, like things you know, dysregulation of some of the factors that we talked about like emp4 or something like that. Then you turn it off.

A

If there's a mutation that turns it off or whether there's a circumstance that turns it down, what is the effect on the asteroid or the embryoid or if the environment's different, if the environment is not as it should be, we talked about the within the embryo environment, which is this secretion of molecules and signaling molecules that guide cells to their faith or the external environment. So if the medium is not as it should be, and.

E

A

A standing for the the environment in the womb, because we don't you know we can't necessarily know exactly what that's like, but we can try to approximate it. So we have these two different types of levels of environment and if they, you know, have aberrations for some reason, then that can affect you. We want to know what that looks like what those effects. Look like this review summarizes the recent developments in the use of gastroites to model congenital abnormalities.

A

There is a generative genetic testing and the current limitations in this emerging field, so you can actually form teratomas and look at some of these congenital anomalies using these models- and this is again, you know this is stuff that is important for medicine and important than other areas as well like drug testing, so I I.

A

This is the review I'm not going to go through the whole thing, I'm just going to say that basically, uh this period of development has been kind of like a black box and now we're kind of making this gray box as the same uh cybernetics, meaning that we know something about right now, but we don't know everything, and so that's work to come.

A

So thank you for paying attention. I hope. You learn something.