There's predictive and then there's hyper-predictive in North Carolina.
Researchers from North Carolina State University have demonstrated that molecular dynamics simulations and machine learning techniques could be integrated to create more accurate computer prediction models. These "hyper-predictive" models could be used to quickly predict which new chemical compounds could be promising drug candidates.
Science Daily: Molecular dynamics, machine learning create 'hyper-predictive' computer models
May our dear Lord forgive them for hyper-destruction of the language with another yet prefix for the sophomore suite of super, mega, ultra.
There's some blather up front about why it's needed and then some ooh wow about how it works.
Fourches and Jeremy Ash, a graduate student in bioinformatics, decided to incorporate the results of molecular dynamics calculations -- all-atom simulations of how a particular compound moves in the binding pocket of a protein -- into prediction models based on machine learning.
"Most models only use the two-dimensional structures of molecules," Fourches says. "But in reality, chemicals are complex three-dimensional objects that move, vibrate and have dynamic intermolecular interactions with the protein once docked in its binding site. You cannot see that if you just look at the 2-D or 3-D structure of a given molecule."
- SD
I'm getting a good measure of ooh wow out of that, mates ... but not enough to forgive the use of hyper-predictive.
We finally get to what's good about it.
"Before computing advances allowed us to simulate this kind of data, it would have taken us six months to simulate one single molecule in the pocket of ERK2. Thanks to GPU acceleration, now it only takes three hours. That is a game changer. I'm hopeful that incorporating data extracted from molecular dynamics into QSAR models will enable a new generation of hyper-predictive models that will help bringing novel, effective drugs onto the market even faster. It's artificial intelligence working for us to discover the drugs of tomorrow."
- SD
Um, no, mates. It doesn't change any game but it does make their process a shitload faster.
Ed: they should use the extra time for learning English and hype reduction
Roger that
Cool science but two demerits for use of mindless cliche (i.e. game-changer) and an attempt to create more (i.e. hyper-predictive).
My ol' Dad was using SIMSCRIPT for modeling on computers in the late 50s and/or early 60s and any modeling is arguably predictive ... and that was almost sixty-five years ago.
Hmm ... maybe SIMSCRIPT came later than that but his computer modeling definitely started in the 50s. That's how he made his bones in genetics.
Note for extreme machine rats: he was running on the SILIAC at CSIRO, Sydney, and that was one of the progeny of the ILIAC machine in America.
Researchers from North Carolina State University have demonstrated that molecular dynamics simulations and machine learning techniques could be integrated to create more accurate computer prediction models. These "hyper-predictive" models could be used to quickly predict which new chemical compounds could be promising drug candidates.
Science Daily: Molecular dynamics, machine learning create 'hyper-predictive' computer models
May our dear Lord forgive them for hyper-destruction of the language with another yet prefix for the sophomore suite of super, mega, ultra.
There's some blather up front about why it's needed and then some ooh wow about how it works.
Fourches and Jeremy Ash, a graduate student in bioinformatics, decided to incorporate the results of molecular dynamics calculations -- all-atom simulations of how a particular compound moves in the binding pocket of a protein -- into prediction models based on machine learning.
"Most models only use the two-dimensional structures of molecules," Fourches says. "But in reality, chemicals are complex three-dimensional objects that move, vibrate and have dynamic intermolecular interactions with the protein once docked in its binding site. You cannot see that if you just look at the 2-D or 3-D structure of a given molecule."
- SD
I'm getting a good measure of ooh wow out of that, mates ... but not enough to forgive the use of hyper-predictive.
We finally get to what's good about it.
"Before computing advances allowed us to simulate this kind of data, it would have taken us six months to simulate one single molecule in the pocket of ERK2. Thanks to GPU acceleration, now it only takes three hours. That is a game changer. I'm hopeful that incorporating data extracted from molecular dynamics into QSAR models will enable a new generation of hyper-predictive models that will help bringing novel, effective drugs onto the market even faster. It's artificial intelligence working for us to discover the drugs of tomorrow."
- SD
Um, no, mates. It doesn't change any game but it does make their process a shitload faster.
Ed: they should use the extra time for learning English and hype reduction
Roger that
Cool science but two demerits for use of mindless cliche (i.e. game-changer) and an attempt to create more (i.e. hyper-predictive).
My ol' Dad was using SIMSCRIPT for modeling on computers in the late 50s and/or early 60s and any modeling is arguably predictive ... and that was almost sixty-five years ago.
Hmm ... maybe SIMSCRIPT came later than that but his computer modeling definitely started in the 50s. That's how he made his bones in genetics.
Note for extreme machine rats: he was running on the SILIAC at CSIRO, Sydney, and that was one of the progeny of the ILIAC machine in America.
No comments:
Post a Comment