Where can I see a progress report of the project? #322
-
|
I would love to know the progress and the current difficulties/blockers |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 4 replies
-
|
Hi, @LifeIsStrange. I hesitate to attempt any kind of summary, generally. The Docker container that's built from this Git repository is one artifact that goes through a slice of the tools and frameworks that were built for, made in association with, or otherwise integrated with OpenWorm over the past several years in terms of modeling and simulation (sibernetic, neuroml, c302, WCON format, owmeta/PyOpenWorm, etc.). It doesn't include some other major things like Geppetto and the robotics project...mostly because it would be awkward to include those things in one Docker container, I think. I won't even pretend to be up on the science side of things. From what I've seen, OpenWorm folks have kept abreast of developments related to C. elegans and shared that work as part of the open science mission of the project. https://openworm.org/science.html has some details, though some parts are probably out-of-date. Most people who are still contributing on the project mostly have their heads down: they make contribution as and when they can, so things aren't too noisy. You can still see a steady stream of commits if you watch the right repositories. You may also follow the organization as a whole since apparently that's a thing now... as far as blockers...I think Andrey and Sergey (Sibernetic) were in need of GPUs to push forward with their work, but had trouble finding them at a reasonable price in Fall of last year. For me (owmeta, sometimes Docker, different Python things), I've just been focused on other things. |
Beta Was this translation helpful? Give feedback.
-
|
Oh Thank you for the answer ! BTW is there video recordings of https://app.genetics-gsa.org/celegans/abstracts21_report/Assignment?lastNameLetter=G ? |
Beta Was this translation helpful? Give feedback.
-
|
Hi @LifeIsStrange. Yes, @mwatts15 has given a good quick summary of the project, much of the recent work has focussed on getting the Docker image working with the latest elements from the OpenWorm Software stack. Much of it is boring but important tasks getting the latest versions of (Python) packages to work together. Some other recent work, focussed on getting more experimental data into the project with which to constrain the model, includes this for electrophysiological data from worm neurons: https://github.com/openworm/WormsenseLab_ASH, and conversion of a dataset with genetically encoded unique colors for each neuron: https://github.com/openworm/NeuroPAL. To be honest, your question highlights the need for a bit of work on our side to more visibly highlight the recent work in the project. One idea we're pursuing is a ~3 monthly newsletter summarising recent developments in the project, which should be a bit more prominent on the website and our communication channels. The question you raise about the scientific advances is a good one, and our progress in this area is held back mainly because we still don't have any central academic funding (though we have got a new stream for donations at OpenCollective). While software contributions from general open source developers are always welcome, we do need PhD/postdocs with knowledge of the area to get involved. An initiative to address this is the OpenWorm Studentship program, aimed specifically at these groups. Getting someone on board to dive deep into a dataset like the one you highlight is exactly the type of task this is intended for. |
Beta Was this translation helpful? Give feedback.
-
|
Thank you, that's a great answer. Actually I put the wrong link in my first comment, the paper I was referring was exactly neuroPAL so it's good to hear that you are working with this dataset! I find it very unfortunate that you lack funding given that understanding how neurons circuits works would be a breakthrough enabling insights that could be backported to the field of artificial intelligence and also has major open problems regarding qualias/consciousness. Brains are so energy and space efficient in terms of compute abilities and accuracy that they in many regards seems to transcend standard computer science and one would even almost believe they bypass the laws of physics... For all those reasons, each being sufficient, you should receive plenty of funds, however in the irationnal word we live in, this won't happen without a campaign fund/marketing. Though you might consider doing a patreon in addition to opencollective ? aside:
I was wondering wether using a different coefficient correlation method that Pearson might reveal higher correlation ? |
Beta Was this translation helpful? Give feedback.
-
|
Thanks, I agree more funding would be great... |
Beta Was this translation helpful? Give feedback.
Hi, @LifeIsStrange. I hesitate to attempt any kind of summary, generally. The Docker container that's built from this Git repository is one artifact that goes through a slice of the tools and frameworks that were built for, made in association with, or otherwise integrated with OpenWorm over the past several years in terms of modeling and simulation (sibernetic, neuroml, c302, WCON format, owmeta/PyOpenWorm, etc.). It doesn't include some other major things like Geppetto and the robotics project...mostly because it would be awkward to include those things in one Docker container, I think.
I won't even pretend to be up on the science side of things. From what I've seen, OpenWorm folks have…