HackerTrans
TopNewTrendsCommentsPastAskShowJobs

Thebroser

no profile record

Submissions

Executive Order on the Safe, Secure, and Trustworthy Development and Use of AI

whitehouse.gov
4 points·by Thebroser·2 वर्ष पहले·0 comments

A universal deep-learning model for zinc finger design

nature.com
1 points·by Thebroser·3 वर्ष पहले·0 comments

Large Language Models Encode Clinical Knowledge

arxiv.org
2 points·by Thebroser·4 वर्ष पहले·0 comments

PhD students face cash crisis with wages that don’t cover living costs

nature.com
236 points·by Thebroser·4 वर्ष पहले·233 comments

comments

Thebroser
·पिछला वर्ष·discuss
The approach they used which is base editing doesn’t actually insert or remove DNA, it actually uses an enzyme to convert one base to another, which is much safer as this doesn’t require a double strand break in DNA: https://blog.addgene.org/single-base-editing-with-crispr
Thebroser
·पिछला वर्ष·discuss
Add gene has a great guide as to what goes on at the molecular level: https://www.addgene.org/guides/crispr/

Essentially you can design an rna molecular that contains a 20 nucleotide long sequence that can target your region of interest, with the caveat that there is a standard recognition sequence proximal to your sequence of interest (PAM sequence)
Thebroser
·2 वर्ष पहले·discuss
I used to love this chrome extension back in the day, unsure if it still works but same idea:

https://ottoplay.tv/
Thebroser
·2 वर्ष पहले·discuss
Home lab, or even be able to sequence your genome at home, privately and securely without risking giving your genetic information to companies like 23andme
Thebroser
·3 वर्ष पहले·discuss
There’s also people pursuing mRNA delivery systems to do this. Might be cheaper than viruses and there’s precedent now with COVID vaccines. https://www.capstantx.com/
Thebroser
·3 वर्ष पहले·discuss
There are research groups that are trying to encode genetic neural networks into cells like the example I have attached, but the neuronal approach from the post does seem to be different here. https://www.nature.com/articles/s41467-022-33288-8
Thebroser
·3 वर्ष पहले·discuss
Would probably assume some level of immunosuppressants might be necessary.
Thebroser
·3 वर्ष पहले·discuss
There are definitely very solid attempts at least to make LLMs that encode biomedical knowledge such as BioGPT which is trained on Pubmed and other domain specific areas. Source: https://arxiv.org/abs/2210.10341
Thebroser
·3 वर्ष पहले·discuss
Spatial Transcriptomics. Being able to see gene expression at a resolution of a handful of cells in a cross section of tissue is revolutionizing disease research. https://www.nature.com/articles/s41592-020-01033-y
Thebroser
·4 वर्ष पहले·discuss
I am always fascinated by the ignorance that is blatantly displayed on HN when it comes to biotech advancements. Not even some actual criticism of the tech, or even the faint understanding that this sort of thing is exploratory research. Wonder why this sort of skepticism is never applied to cutting edge AI research whose ethical ramifications are much more severe and which is already being utilized by bad actors to facilitate genocide [1].

[1] https://wires.onlinelibrary.wiley.com/doi/abs/10.1002/widm.1...
Thebroser
·4 वर्ष पहले·discuss
“Animal research has shown that natural cochlear hair cell regeneration and resultant hearing restoration is very real” - Although exciting, we also need to remember that translatability rates from animal models to humans is notoriously low, usually in the single digits for most therapeutic areas. This is some cool tech, but just wanted to point it out that success in animal models =/= we will eventually get to see the realized treatment.
Thebroser
·4 वर्ष पहले·discuss
Thanks a ton for your reply! I wish you all the best, tons of cool innovation to be done here.
Thebroser
·4 वर्ष पहले·discuss
Hey! Bioengineering background here, how do you all embed the cells within the scaffold? Is your approach similar to other bioprinting approaches (initial cells seeded via extrusion, inkjet, or laser assisted deposition)? At first I would think this could be done with some sort of printable collagen scaffold but I'm curious for what your approach would be. In addition, as someone that is involved in therapeutics, I'm not too well versed in this space but was curious hearing from your perspective: How close would you say we are to being able to use FBS free media to culture meat at scale?
Thebroser
·4 वर्ष पहले·discuss
If you pick up an intro college level bio book that might be enough to start out with. My university publishes this so it would be a nice free start -> https://openstax.org/details/books/biology-2e

After that, you could maybe get interested into a anatomy/physiology or systems physiology textbook.

For more specialized content maybe NCBI StatPearls -> https://www.ncbi.nlm.nih.gov/books/NBK430685/

Note that due to the ever changing landscape of biology, only way to truly ever "be an expert" is to actively participate in the life sciences and keep up with contemporary research articles.
Thebroser
·4 वर्ष पहले·discuss
The thing about these developmental conditions is that in theory there a lot of different treatment modalities that can be pursued at the embryonic level that could address these conditions (viral or non viral gene therapy, CRISPR based base deletions/insertions of the correct allele, etc.). However, once embryonic development has occurred and these conditions have manifested, it would be very difficult to alleviate them. Note that this is very different from something like sickle cell anemia. As a metaphor, think of the human body as a brick house. Using state of the art biotech to treat sickle cell anemia would be like trying to replace a leaky pipe hidden under concrete in the house. Yes it may be difficult but it is doable as it’s a fairy isolated issue. Trying to fix developmental diseases like KS would be like trying to replace every inch of the mortar in the brick house. It would be painstakingly difficult to do it right without the whole house collapsing. In summary, the tech you are referencing is basically limited to pre-developmental organisms, it needs decades or centuries to go before we may feasibly perform large scale genetic engineering that can undo post-developmental physiology.

Next, we have to understand that there are serious ethical considerations on treatments that target embryos. Biology is still very much a black box field. It is near impossible trying to factor in all the parameters in this problem space, especially when you are altering genes at a highly regulated stage of human development. Millions of model organisms like mice and non human primates are used to test therapies, but would we be willing to do the same to human embryos at such a large scale? I would hardly imagine an IRB approving research like this in my lifetime in the U.S.
Thebroser
·5 वर्ष पहले·discuss
Yes! The Open Insulin foundation not only has volunteers with a biomedical background, we have volunteers from many other backgrounds including lawyers that help us try and challenge existing patents. Again, others have said that manpower might be better invested in openly challenging the monopoly that U.S. companies have on some of these patents in court or by appealing to our congressmen. However, I believe that there’s no reason not to do both in parallel. My scientific training can only take me so far and i’ll happily help a project like this if it can open up a new source of affordable insulin in this country, but believe me, I know that at the root this is a legislative issue.
Thebroser
·5 वर्ष पहले·discuss
There are crazy patents throughout the entirety of the insulin production pipeline. Some of these being extremely absurd such as patenting the pH of a buffer for specific protocols! I volunteer for the Open Insulin Foundation on/off when not too focused on my research project and it's basically how can we come up with a roadmap to making this decentralized insulin without violating any existing patents. An added layer of difficulty is like other have mentioned, engineering Ecoli to produce the homolog as well as developing from scratch the devices for extraction and purification (think hplc). Again, these need to be made from scratch as to not violate any patents. At least in the U.S., this is why there is no competition in the market. There is also a significant price barrier as you need to bring up your production to GMP standards which means you need a state of the art clean room and multiple roles to supervise the production of the product every step of the way.