SciTech #ScienceSunday Digest - 47/2016.
SciTech #ScienceSunday Digest - 47/2016.
Permalink here: http://www.scitechdigest.net/2016/11/compartmentalised-gene-circuits-crispr.html
Compartmentalised gene circuits, CRISPR human trial, Google’s machine learning, Parabiosis twist, Carbon nanotube terahertz scanner, Automated drug discovery, DNA origami muscles, Magnetic hand tracking, Photonic neuromorphic computing, Nanoantenna optical switches.
1. Compartmentalised Synthetic Gene Circuits
The utility and power of synthetic biology has been boosted with new modular tools that involve placing different genetic circuits into separate liposomes within the modified cells http://news.mit.edu/2016/synthetic-cells-isolate-genetic-circuits-1114. This solves the potential problem of too many genetic circuits interfering with each other, while allowing the same circuit to be used in a different way at the same time. As a demonstration a circuit in one liposome reacted to a drug by releasing another molecule from its liposome, which entered a second liposome and genetic circuit that responded by producing a light-emitting protein. Other triggers include forcing the liposomes to merge and combine contents. I think of this system as producing basic little custom-engineered and custom-programmed nanofactories in cells. Such modularity makes this a very powerful platform for pushing synthetic biology applications forward.
2. CRISPR Therapy in First Human Trial
A Chinese group has introduced a CRISPR-based therapy into humans as part of a human trial for the first time http://www.nature.com/news/crispr-gene-editing-tested-in-a-person-for-the-first-time-1.20988. In this case the CRISPR modification took place in immune cells extracted from the patient, disabling the PD-1 gene that inhibits cellular immune response, and then these cells were reintroduced back into patients with lung cancer in the hope that the unhindered immune cells would defeat it. 2017 will see an number of other CRISPR human trials begin, most of which will be targeting various cancers. As CRISPR tools get better I’m expecting this space to explode.
3. Google Machine Learning Advances
First, Google demonstrates RAISR, a machine learning based tool that upscales low resolution images to high resolution copies, which might improve the viewability of low resolution images or otherwise preserve bandwidth http://www.androidpolice.com/2016/11/14/google-announces-raisr-method-upscaling-images-machine-learning/. Second, better machine learning is powering the latest iteration of Google Translate, which can now translate whole sentences at a time rather than piece by piece, and resulting in much smoother grammatically correct translations https://blog.google/products/translate/found-translation-more-accurate-fluent-sentences-google-translate/.
4. Surprising Twist with Parabiosis
Parabiosis is the term for connecting the circulatory systems of two animals together, and when done with an old and young mouse the older mouse shows signs of regeneration to a more youthful state. In an interesting twist, the blood plasma from young (18 year old) humans was taken and injected into old mice (1 year old, ~50 year human equivalent) and this also rejuvenated the old mice to a more youthful state as measured by movement, memory function, and increased neurogenesis https://www.newscientist.com/article/2112829-blood-from-human-teens-rejuvenates-body-and-brains-of-old-mice/. Interesting both for the fact that an 18 year old human’s blood rejuvenated a one year old mouse, and also for the promise this general procedure and the isolation of specific factors has for human health applications.
5. Portable Carbon Nanotube Terahertz Scanner
A portable, flexible, wearable, terahertz scanner has been created from thin films of carbon nanotubes http://spectrum.ieee.org/nanoclast/semiconductors/nanotechnology/flexible-portable-terahertz-scanner-made-from-carbon-nanotubes2. This effectively amounts to a flexible terahertz camera that in tests was able to detect and image a wide band of terahertz rays. Given terahertz light passes through many materials the promise has always been to use such devices for security as part of high-resolution non-invasive imaging to detect hidden objects.
6. Automated Drug Discovery
First, iPANDA is a machine learning system for discovering new drugs from gene expression data and which outputs the pharmacological properties of new potential drugs and biomarkers for further development https://www.eurekalert.org/pub_releases/2016-11/imi-ian111616.php. Second, another machine learning system was being used to study antimicrobial compounds and helped discover general peptides and protein features that can cross cell membranes, thus not only enabling the design of different and better antimicrobial peptides but also peptides that can ferry drugs and other molecules into cells http://newsroom.ucla.edu/releases/computers-learn-to-recognize-molecules-that-can-enter-cells.
7. DNA Origami Muscles
Nanoscale muscles have been built from gold nanoparticles linked in sheets by different single strand DNA sequences and with these sheets stacked in many layers https://news.upenn.edu/news/penn-engineers-make-nanoscale-muscles-powered-dna. By introducing different complementary strands induces the formation of double-stranded DNA bridges in precise locations, which are different lengths to single strands, and this causes the stacked sheets to twist and flex and roll up in controllable shapes. This controllable flexing behaviour could be used in nanoscale diagnostics, for example, recognising a certain type of RNA in the cell by changing the spacing between layers in the sheet and so interacting with light differently to provide a signal of the event.
8. VR Magnetic Hand Tracking
Ommo is developing a system to provide millimeter resolution gesture tracking in virtual reality environments magnetic fields and sensors http://spectrum.ieee.org/view-from-the-valley/at-work/start-ups/for-precise-hand-tracking-in-virtual-reality-start-with-a-magnetic-field. This includes a small magnetic field generator that is worn in the pocket and sensors on key points of a glove worn on the hands. Developer prototypes are expected next year. I’ve been playing with Daydream View lately and it would be pretty amazing to have this type of control and gesture interface in these environments.
9. Photonic Neuromorphic Computing
Neuromorphic computer hardware has taken a big step with the development of the first photonic neuromorphic chip for processing neural networks and deep learning algorithms at ultrafast speeds https://www.technologyreview.com/s/602938/worlds-first-photonic-neural-network-unveiled/. For certain types of information processing, types of recurrent neural networks for example, the photonic circuits have been demonstrated to be 2,000 times faster than conventional processing. The photonic circuits are comprised of nodes that respond in a similar manner to a neuron and consist of circular waveguides that trap light, which when released modulates the output of an associated laser.
10. Nanoantenna Switches for Optical Computing
On the topic of photonic chips, new nanoantennas made from silicon nanoparticles efficiently switch the direction of incoming light depending on the intensity of that light http://spectrum.ieee.org/nanoclast/semiconductors/nanotechnology/nanoantenna-changes-direction-of-light-and-the-prospects-of-optical-computing. Low intensity light is unaffected, while high intensity light generates a type of electron plasma around the silicon nanoparticles, resulting in a significant change to the refractive index of the nanoantennas, and so bending the light in a measurable and different way. These nanoantennas can support data rates up to 250 gigabits per second and offer a great platform for developing optical computing applications and chips in future.
SciTech Tip Jar: http://www.scitechdigest.net/p/donate.html
Permalink here: http://www.scitechdigest.net/2016/11/compartmentalised-gene-circuits-crispr.html
Compartmentalised gene circuits, CRISPR human trial, Google’s machine learning, Parabiosis twist, Carbon nanotube terahertz scanner, Automated drug discovery, DNA origami muscles, Magnetic hand tracking, Photonic neuromorphic computing, Nanoantenna optical switches.
1. Compartmentalised Synthetic Gene Circuits
The utility and power of synthetic biology has been boosted with new modular tools that involve placing different genetic circuits into separate liposomes within the modified cells http://news.mit.edu/2016/synthetic-cells-isolate-genetic-circuits-1114. This solves the potential problem of too many genetic circuits interfering with each other, while allowing the same circuit to be used in a different way at the same time. As a demonstration a circuit in one liposome reacted to a drug by releasing another molecule from its liposome, which entered a second liposome and genetic circuit that responded by producing a light-emitting protein. Other triggers include forcing the liposomes to merge and combine contents. I think of this system as producing basic little custom-engineered and custom-programmed nanofactories in cells. Such modularity makes this a very powerful platform for pushing synthetic biology applications forward.
2. CRISPR Therapy in First Human Trial
A Chinese group has introduced a CRISPR-based therapy into humans as part of a human trial for the first time http://www.nature.com/news/crispr-gene-editing-tested-in-a-person-for-the-first-time-1.20988. In this case the CRISPR modification took place in immune cells extracted from the patient, disabling the PD-1 gene that inhibits cellular immune response, and then these cells were reintroduced back into patients with lung cancer in the hope that the unhindered immune cells would defeat it. 2017 will see an number of other CRISPR human trials begin, most of which will be targeting various cancers. As CRISPR tools get better I’m expecting this space to explode.
3. Google Machine Learning Advances
First, Google demonstrates RAISR, a machine learning based tool that upscales low resolution images to high resolution copies, which might improve the viewability of low resolution images or otherwise preserve bandwidth http://www.androidpolice.com/2016/11/14/google-announces-raisr-method-upscaling-images-machine-learning/. Second, better machine learning is powering the latest iteration of Google Translate, which can now translate whole sentences at a time rather than piece by piece, and resulting in much smoother grammatically correct translations https://blog.google/products/translate/found-translation-more-accurate-fluent-sentences-google-translate/.
4. Surprising Twist with Parabiosis
Parabiosis is the term for connecting the circulatory systems of two animals together, and when done with an old and young mouse the older mouse shows signs of regeneration to a more youthful state. In an interesting twist, the blood plasma from young (18 year old) humans was taken and injected into old mice (1 year old, ~50 year human equivalent) and this also rejuvenated the old mice to a more youthful state as measured by movement, memory function, and increased neurogenesis https://www.newscientist.com/article/2112829-blood-from-human-teens-rejuvenates-body-and-brains-of-old-mice/. Interesting both for the fact that an 18 year old human’s blood rejuvenated a one year old mouse, and also for the promise this general procedure and the isolation of specific factors has for human health applications.
5. Portable Carbon Nanotube Terahertz Scanner
A portable, flexible, wearable, terahertz scanner has been created from thin films of carbon nanotubes http://spectrum.ieee.org/nanoclast/semiconductors/nanotechnology/flexible-portable-terahertz-scanner-made-from-carbon-nanotubes2. This effectively amounts to a flexible terahertz camera that in tests was able to detect and image a wide band of terahertz rays. Given terahertz light passes through many materials the promise has always been to use such devices for security as part of high-resolution non-invasive imaging to detect hidden objects.
6. Automated Drug Discovery
First, iPANDA is a machine learning system for discovering new drugs from gene expression data and which outputs the pharmacological properties of new potential drugs and biomarkers for further development https://www.eurekalert.org/pub_releases/2016-11/imi-ian111616.php. Second, another machine learning system was being used to study antimicrobial compounds and helped discover general peptides and protein features that can cross cell membranes, thus not only enabling the design of different and better antimicrobial peptides but also peptides that can ferry drugs and other molecules into cells http://newsroom.ucla.edu/releases/computers-learn-to-recognize-molecules-that-can-enter-cells.
7. DNA Origami Muscles
Nanoscale muscles have been built from gold nanoparticles linked in sheets by different single strand DNA sequences and with these sheets stacked in many layers https://news.upenn.edu/news/penn-engineers-make-nanoscale-muscles-powered-dna. By introducing different complementary strands induces the formation of double-stranded DNA bridges in precise locations, which are different lengths to single strands, and this causes the stacked sheets to twist and flex and roll up in controllable shapes. This controllable flexing behaviour could be used in nanoscale diagnostics, for example, recognising a certain type of RNA in the cell by changing the spacing between layers in the sheet and so interacting with light differently to provide a signal of the event.
8. VR Magnetic Hand Tracking
Ommo is developing a system to provide millimeter resolution gesture tracking in virtual reality environments magnetic fields and sensors http://spectrum.ieee.org/view-from-the-valley/at-work/start-ups/for-precise-hand-tracking-in-virtual-reality-start-with-a-magnetic-field. This includes a small magnetic field generator that is worn in the pocket and sensors on key points of a glove worn on the hands. Developer prototypes are expected next year. I’ve been playing with Daydream View lately and it would be pretty amazing to have this type of control and gesture interface in these environments.
9. Photonic Neuromorphic Computing
Neuromorphic computer hardware has taken a big step with the development of the first photonic neuromorphic chip for processing neural networks and deep learning algorithms at ultrafast speeds https://www.technologyreview.com/s/602938/worlds-first-photonic-neural-network-unveiled/. For certain types of information processing, types of recurrent neural networks for example, the photonic circuits have been demonstrated to be 2,000 times faster than conventional processing. The photonic circuits are comprised of nodes that respond in a similar manner to a neuron and consist of circular waveguides that trap light, which when released modulates the output of an associated laser.
10. Nanoantenna Switches for Optical Computing
On the topic of photonic chips, new nanoantennas made from silicon nanoparticles efficiently switch the direction of incoming light depending on the intensity of that light http://spectrum.ieee.org/nanoclast/semiconductors/nanotechnology/nanoantenna-changes-direction-of-light-and-the-prospects-of-optical-computing. Low intensity light is unaffected, while high intensity light generates a type of electron plasma around the silicon nanoparticles, resulting in a significant change to the refractive index of the nanoantennas, and so bending the light in a measurable and different way. These nanoantennas can support data rates up to 250 gigabits per second and offer a great platform for developing optical computing applications and chips in future.
SciTech Tip Jar: http://www.scitechdigest.net/p/donate.html
Comments
Post a Comment