Is Google DeepMind’s Protein Structure Prediction Model a Game Changer for Scientists?

Craig: deepmind. now we talked about this a while ago, around meta's tool that they found all these new, proteins. and yep. it was great. but now they've open sourced it. so now the scientists can actually get their hands on the data and use the tool. what do you think about this?

Jon: yeah. so, google's parent company, deepmind or sorry. deepmind, which is a subsidiary of alphabet. so they released alphafold three, which represents an advancement in the field of structural biology. so it can now not only predict the structure of individual proteins, but also how proteins interact with other molecules in call including dna, rna, and small molecules like drugs. so this is really cool because the broader in, input the broader application of this is that you can now model a wider range of biological molecules. so now it makes it a more versatile tool for researchers across multiple different scientific disciplines. so i think this is one of the biggest advancements as far as when it comes to this type of research, in order to, enhance precision predicting protein to protein interactions, which is more complex kind of protein, which is helpful for more large complex proteins. right?

Craig: yeah. and so just to get a handle on this a little bit, do you understand for the audience to understand that protein structures could lead to cure for diseases like alzheimer's or even cancer? imagine having a super smart friend who could predict how your little league team would perform against the pros. this is exactly what alphafold is doing, but for proteins. now let's unfold. let's talk about the story a little bit more. like, what you can do what this product can do is leaps and bounds over, what the scientists had as a tool before and understanding how proteins fold into functional forms is a bit like trying to solve a rubik's cube in the dark. the traditional methods like x-ray, crystallography and nmr spectro spectro stress the spectroscopy? spectroscopy.

Jon: oh.

Craig: for you effective, but, oh, so slow and resource hungry. you know, this came along and, you know, it looks at the existing protein data of over a hundred and seventy thousand structures and tells scientists, here's what you're working with. it's like having a cheat sheet for an exam that used to take years to study for. i'd say it's, like, leaps and bounds ahead.

Jon: yeah. this is cool because what it does is it kind of levels the playing field. and because as it becomes more accessible, it allows access for researchers worldwide, which enables smaller labs to contribute to this cutting edge research, which is really good. and i think we actually talked about that previous, in our previous talk about the smaller labs being able to access things like this to, to help advance the research. so, there there obviously is there obviously is a lot more accessibility, but they do raise concerns about data ownership, privacy. and, again, like we always talk, the potential for misuse of this technology.

Craig: yeah. the implications are huge. we're, you know, we're talking about speeding up drug discovery processes, enhancing knowledge about diseases, and even redesigning proteins. you know, picture a pharmaceutical company waltzing through their r and d processes thanks to alpha fold predictions. suddenly, they're tossing around terms like accelerated pathways for alzheimer's treatments as if they're ordering coffee because now, you know, they have this tool that understands these interactions far better than they have ever had before.

Jon: yeah. and i could be wrong in this, but i think they're starting to say that also is it alzheimer's that is diabetes type three?

Craig: they're not they're starting to say it's diabetes related. absolutely.

Jon: yeah. so, it's cool. i like this because if you think about it this way, what they could do is what would take years to potential to potentially solve, it could take weeks to potentially solve. right? and so the other thing too and i i think we covered this where canada had come up with, a research or, maybe a university in canada had come up with, i think, how many different new proteins or new interactions or something like that? i can't remember. jesus. that was a while ago.

Craig: it was a while ago, but it was a lot. like, it did what it would take years to do in hours. and, you know, that's how forward leaping we are. and look as we look to the future, the wonders of machine learning continue to unfold, promising a world where biological research is faster, smarter, and even more insightful. with alphafold leading the way, the chemistry world's just begun to mix. so here's the question for you, our listeners. how do you see these advances in ai like alphafold shaping the future of health care and medicine? and is this just the beginning of the revolution in scientific discovery? jump on our social media channels and leave a comment in the comments below. we'd love to hear your thoughts.

Jon: yeah. thanks for the interaction, colby. you've been great so far.

Craig: alright. i think we'll leave it at that, and then i'll ask you, what do you got going on, john?

Jon: yeah. so, next week, i'm excited. there's, i'm going to tech the halls, which is an interesting, mnp event that's gonna be fun on thursday. and, also at, the toronto business development centre, they're hosting, a venture day on the same day on thursday. so, i that's what i have going on, which is exciting. so i get two birds with one stone in one day, which is great because i've been to toronto in a while. so how about you, craig?

Craig: well, this will be coming out early december, december fifth, and depending on the art on what you're listening to. so most of the stuff will be passed, but we probably still have some of our hangover from our black friday deals, coming out. so check out toronto star's website for that where you'll get deals on upcoming courses, on, our coaching programs and mentoring and several other things. so there's some deals going on there. if you're looking to start a business or grow your business, now is the best time to jump in on those things. and we're looking at some new venues in the new year for start up investor drinks, and, you know, i'm always doing interviews, talking to new start ups as we start to wind down the year.

Jon: yeah. this is great.

Craig: and, you help people with non dilutive funding. where should they go?

Jon: yeah. come find me on linkedin, john irwin. just so the just, i don't know when this comes out, but i'm sure it'll still be open. canada summer jobs is open. so for those looking to hire interns, for the summer, apply now. you get fifty percent of their minimum wage rate covered. so come find me. it's it's an easy one to do. you can do it on your own, but, if you need some help with your business, coaching, your start up. craig, did you not you working on something, or did or how do you help people?

Craig: well, we do mentoring programs. we do coaching. i have different coaching packages. you can go to start up coach dot c a and book a fifteen minute discovery call and see if that's something that you're interested. we also, are launching our course in mid january, for launching your business in a weekend. if you've been struggling and not taking it forward, you're not sure about your idea, you're not sure about a lot of things, we're gonna take you from that not sure place to, being selling stuff at the end of the weekend and ready to do it. you know? it's amazing what we're gonna be doing. check out startup coach dot c a for some of that. again, we have coaching programs. we have mentoring programs. we have online educational stuff. and if you just need some help, send me an email at craig at toronto starts dot com, and i'll point you in the right direction. or, look me up on linkedin as well. and thanks everyone for listening.

Jon: thanks everyone.