Matias Kaplan
Research Subjects
Matias Kaplan has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Christina D. Smolke
Department of Bioengineering, Stanford University, Stanford, CA 94305
4 shared publications - Brent Townshend
Stanford University
2 shared publications - Jennifer A. Doudna
Department of Molecular and Cell Biology, University of California, Berkeley, Berkeley, CA, USA
2 shared publications Samuel H. Sternberg
Department of Chemistry, University of California, Berkeley, California 94720, USA
2 shared publications- Adrian L. Sanborn
Department of Computer Science, Stanford University
1 shared publication Akshay Tambe
Atomic AI
1 shared publication
Latest Publications
- Open Access
ATOM-1: A Foundation Model for RNA Structure and Function Built on Chemical Mapping Data
RNA-based medicines and RNA-targeting drugs are emerging as promising new approaches for treating disease. Optimizing these therapeutics by naive experimental screening is a time-consuming and expensive process, while rational design requires an accurate understanding of the structure and function of RNA. To address this design challenge, we present ATOM-1, the first RNA foundation model trained on chemical mapping data, enabled by data collection strategies purposely developed for machine learning training. Using small probe neural networks on top of ATOM-1 embeddings, we demonstrate that this model has developed rich internal representations of RNA. Trained on limited amounts of additional data, these small networks achieve state-of-the-art accuracy on key RNA prediction tasks, suggesting that this approach can enable the design of therapies across the RNA landscape.
Open AccessATOM-1: A Foundation Model for RNA Structure and Function Built on Chemical Mapping Data
RNA-based medicines and RNA-targeting drugs are emerging as promising new approaches for treating disease. Optimizing these therapeutics by naive experimental screening is a time-consuming and expensive process, while rational design requires an accurate understanding of the structure and function of RNA. To address this design challenge, we present ATOM-1, the first RNA foundation model trained on chemical mapping data, enabled by data collection strategies purposely developed for machine learning training. Using small probe neural networks on top of ATOM-1 embeddings, we demonstrate that this model has developed rich internal representations of RNA. Trained on limited amounts of additional data, these small networks achieve state-of-the-art accuracy on key RNA prediction tasks, suggesting that this approach can enable the design of therapies across the RNA landscape.
- Open Access
- 15 September 2022
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 17 (9) , e0273381
Highly multiplexed selection of RNA aptamers against a small molecule library
Applications of synthetic biology spanning human health, industrial bioproduction, and ecosystem monitoring often require small molecule sensing capabilities, typically in the form of genetically encoded small molecule biosensors. Critical to the deployment of greater numbers of these systems are methods that support the rapid development of such biosensors against a broad range of small molecule targets. Here, we use a previously developed method for selection of RNA biosensors against unmodified small molecules (DRIVER) to perform a selection against a densely multiplexed mixture of small molecules, representative of those employed in high-throughput drug screening. Using a mixture of 5,120 target compounds randomly sampled from a large diversity drug screening library, we performed a 95-round selection and then analyzed the enriched RNA biosensor library using next generation sequencing (NGS). From our analysis, we identified RNA biosensors with at least 2-fold change in signal in the presence of at least 217 distinct target compounds with sensitivities down to 25 nM. Although many of these biosensors respond to multiple targets, clustering analysis indicated at least 150 different small-molecule sensing patterns. We also built a classifier that was able to predict whether the biosensors would respond to a new compound with an average precision of 0.82. Since the target compound library was designed to be representative of larger diversity compound libraries, we expect that the described approach can be used with similar compound libraries to identify aptamers against other small molecules with a similar success rate. The new RNA biosensors (or their component aptamers) described in this work can be further optimized and used in applications such as biosensing, gene control, or enzyme evolution. In addition, the data presented here provide an expanded compendium of new RNA aptamers compared to the 82 small molecule RNA aptamers published in the literature, allowing further bioinformatic analyses of the general classes of small molecules for which RNA aptamers can be found.
- 15 September 2022
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 17 (9) , e0273381
Open AccessHighly multiplexed selection of RNA aptamers against a small molecule library
Applications of synthetic biology spanning human health, industrial bioproduction, and ecosystem monitoring often require small molecule sensing capabilities, typically in the form of genetically encoded small molecule biosensors. Critical to the deployment of greater numbers of these systems are methods that support the rapid development of such biosensors against a broad range of small molecule targets. Here, we use a previously developed method for selection of RNA biosensors against unmodified small molecules (DRIVER) to perform a selection against a densely multiplexed mixture of small molecules, representative of those employed in high-throughput drug screening. Using a mixture of 5,120 target compounds randomly sampled from a large diversity drug screening library, we performed a 95-round selection and then analyzed the enriched RNA biosensor library using next generation sequencing (NGS). From our analysis, we identified RNA biosensors with at least 2-fold change in signal in the presence of at least 217 distinct target compounds with sensitivities down to 25 nM. Although many of these biosensors respond to multiple targets, clustering analysis indicated at least 150 different small-molecule sensing patterns. We also built a classifier that was able to predict whether the biosensors would respond to a new compound with an average precision of 0.82. Since the target compound library was designed to be representative of larger diversity compound libraries, we expect that the described approach can be used with similar compound libraries to identify aptamers against other small molecules with a similar success rate. The new RNA biosensors (or their component aptamers) described in this work can be further optimized and used in applications such as biosensing, gene control, or enzyme evolution. In addition, the data presented here provide an expanded compendium of new RNA aptamers compared to the 82 small molecule RNA aptamers published in the literature, allowing further bioinformatic analyses of the general classes of small molecules for which RNA aptamers can be found.
- Open Access
Highly multiplexed selection of RNA aptamers against a small molecule library
Applications of synthetic biology spanning human health, industrial bioproduction, and ecosystem monitoring often require small molecule sensing capabilities, typically in the form of genetically encoded small molecule biosensors. Critical to the deployment of greater numbers of these systems are methods that support the rapid development of such biosensors against a broad range of small molecule targets. Here, we use a previously developed method for selection of RNA biosensors against unmodified small molecules (DRIVER) to perform a selection against a densely multiplexed mixture of small molecules, representative of those employed in high-throughput drug screening. Using a mixture of 5,120 target compounds randomly sampled from a large diversity drug screening library, we performed a 95-round selection and then analyzed the enriched RNA biosensor library using next generation sequencing (NGS). From our analysis, we identified RNA biosensors with at least 2-fold change in signal in the presence of at least 217 distinct target compounds with sensitivities down to 25 nM. Although many of these biosensors respond to multiple targets, clustering analysis indicated at least 150 different small-molecule sensing patterns. We also built a classifier that was able to predict whether the biosensors would respond to a new compound with an average precision of 0.82. Since the target compound library was designed to be representative of larger diversity compound libraries, we expect that the described approach can be used with similar compound libraries to identify aptamers against other small molecules with a similar success rate. The new RNA biosensors (or their component aptamers) described in this work can be further optimized and used in applications such as biosensing, gene control, or enzyme evolution. In addition, the data presented here provide an expanded compendium of new RNA aptamers compared to the 82 small molecule RNA aptamers published in the literature, allowing further bioinformatic analyses of the general classes of small molecules for which RNA aptamers can be found.
Open AccessHighly multiplexed selection of RNA aptamers against a small molecule library
Applications of synthetic biology spanning human health, industrial bioproduction, and ecosystem monitoring often require small molecule sensing capabilities, typically in the form of genetically encoded small molecule biosensors. Critical to the deployment of greater numbers of these systems are methods that support the rapid development of such biosensors against a broad range of small molecule targets. Here, we use a previously developed method for selection of RNA biosensors against unmodified small molecules (DRIVER) to perform a selection against a densely multiplexed mixture of small molecules, representative of those employed in high-throughput drug screening. Using a mixture of 5,120 target compounds randomly sampled from a large diversity drug screening library, we performed a 95-round selection and then analyzed the enriched RNA biosensor library using next generation sequencing (NGS). From our analysis, we identified RNA biosensors with at least 2-fold change in signal in the presence of at least 217 distinct target compounds with sensitivities down to 25 nM. Although many of these biosensors respond to multiple targets, clustering analysis indicated at least 150 different small-molecule sensing patterns. We also built a classifier that was able to predict whether the biosensors would respond to a new compound with an average precision of 0.82. Since the target compound library was designed to be representative of larger diversity compound libraries, we expect that the described approach can be used with similar compound libraries to identify aptamers against other small molecules with a similar success rate. The new RNA biosensors (or their component aptamers) described in this work can be further optimized and used in applications such as biosensing, gene control, or enzyme evolution. In addition, the data presented here provide an expanded compendium of new RNA aptamers compared to the 82 small molecule RNA aptamers published in the literature, allowing further bioinformatic analyses of the general classes of small molecules for which RNA aptamers can be found.
- Open Access
- 4 January 2022
- Published by Springer Nature in Nature Reviews Genetics
- Vol. 23 (4) , 215-228
Engineering synthetic RNA devices for cell control
The versatility of RNA in sensing and interacting with small molecules, proteins and other nucleic acids while encoding genetic instructions for protein translation makes it a powerful substrate for engineering biological systems. RNA devices integrate cellular information sensing, processing and actuation of specific signals into defined functions and have yielded programmable biological systems and novel therapeutics of increasing sophistication. However, challenges centred on expanding the range of analytes that can be sensed and adding new mechanisms of action have hindered the full realization of the field’s promise. Here, we describe recent advances that address these limitations and point to a significant maturation of synthetic RNA-based devices.
- 4 January 2022
- Published by Springer Nature in Nature Reviews Genetics
- Vol. 23 (4) , 215-228
Open AccessEngineering synthetic RNA devices for cell control
The versatility of RNA in sensing and interacting with small molecules, proteins and other nucleic acids while encoding genetic instructions for protein translation makes it a powerful substrate for engineering biological systems. RNA devices integrate cellular information sensing, processing and actuation of specific signals into defined functions and have yielded programmable biological systems and novel therapeutics of increasing sophistication. However, challenges centred on expanding the range of analytes that can be sensed and adding new mechanisms of action have hindered the full realization of the field’s promise. Here, we describe recent advances that address these limitations and point to a significant maturation of synthetic RNA-based devices.
- Open Access
- 23 September 2019
- Published by Springer Nature in Nature Communications
- Vol. 10 (1) , 1-16
Massively parallel RNA device engineering in mammalian cells with RNA-Seq
Synthetic RNA-based genetic devices dynamically control a wide range of gene-regulatory processes across diverse cell types. However, the limited throughput of quantitative assays in mammalian cells has hindered fast iteration and interrogation of sequence space needed to identify new RNA devices. Here we report developing a quantitative, rapid and high-throughput mammalian cell-based RNA-Seq assay to efficiently engineer RNA devices. We identify new ribozyme-based RNA devices that respond to theophylline, hypoxanthine, cyclic-di-GMP, and folinic acid from libraries of ~22,700 sequences in total. The small molecule responsive devices exhibit low basal expression and high activation ratios, significantly expanding our toolset of highly functional ribozyme switches. The large datasets obtained further provide conserved sequence and structure motifs that may be used for rationally guided design. The RNA-Seq approach offers a generally applicable strategy for developing broad classes of RNA devices, thereby advancing the engineering of genetic devices for mammalian systems.
- 23 September 2019
- Published by Springer Nature in Nature Communications
- Vol. 10 (1) , 1-16
Open AccessMassively parallel RNA device engineering in mammalian cells with RNA-Seq
Synthetic RNA-based genetic devices dynamically control a wide range of gene-regulatory processes across diverse cell types. However, the limited throughput of quantitative assays in mammalian cells has hindered fast iteration and interrogation of sequence space needed to identify new RNA devices. Here we report developing a quantitative, rapid and high-throughput mammalian cell-based RNA-Seq assay to efficiently engineer RNA devices. We identify new ribozyme-based RNA devices that respond to theophylline, hypoxanthine, cyclic-di-GMP, and folinic acid from libraries of ~22,700 sequences in total. The small molecule responsive devices exhibit low basal expression and high activation ratios, significantly expanding our toolset of highly functional ribozyme switches. The large datasets obtained further provide conserved sequence and structure motifs that may be used for rationally guided design. The RNA-Seq approach offers a generally applicable strategy for developing broad classes of RNA devices, thereby advancing the engineering of genetic devices for mammalian systems.
- Open Access
- 28 October 2015
- Published by Springer Nature in Nature
- Vol. 527 (7576) , 110-113
Conformational control of DNA target cleavage by CRISPR–Cas9
A fluorescence-based approach defines new features of Cas9 that control the specificity of RNA-guided DNA cleavage in CRISPR genome editing technology. The CRISPR genome editing technology now used widely in mutant analysis in a variety of organisms is only a few years old, and the technology is still being optimized to enhance its specificity and efficiency. The central catalytic activity, RNA-guided cleavage that is directed against the genomic DNA, is carried out by the Cas9 endonuclease. Jennifer Doudna and colleagues use a fluorescence-based approach to define new features of Cas9 that control the specificity of RNA-guided DNA cleavage in CRISPR genome-editing technology. Cas9 is an RNA-guided DNA endonuclease that targets foreign DNA for destruction as part of a bacterial adaptive immune system mediated by clustered regularly interspaced short palindromic repeats (CRISPR)1,2. Together with single-guide RNAs3, Cas9 also functions as a powerful genome engineering tool in plants and animals4,5,6, and efforts are underway to increase the efficiency and specificity of DNA targeting for potential therapeutic applications7,8. Studies of off-target effects have shown that DNA binding is far more promiscuous than DNA cleavage9,10,11, yet the molecular cues that govern strand scission have not been elucidated. Here we show that the conformational state of the HNH nuclease domain directly controls DNA cleavage activity. Using intramolecular Förster resonance energy transfer experiments to detect relative orientations of the Cas9 catalytic domains when associated with on- and off-target DNA, we find that DNA cleavage efficiencies scale with the extent to which the HNH domain samples an activated conformation. We furthermore uncover a surprising mode of allosteric communication that ensures concerted firing of both Cas9 nuclease domains. Our results highlight a proofreading mechanism beyond initial protospacer adjacent motif (PAM) recognition12 and RNA–DNA base-pairing3 that serves as a final specificity checkpoint before DNA double-strand break formation.
- 28 October 2015
- Published by Springer Nature in Nature
- Vol. 527 (7576) , 110-113
Open AccessConformational control of DNA target cleavage by CRISPR–Cas9
A fluorescence-based approach defines new features of Cas9 that control the specificity of RNA-guided DNA cleavage in CRISPR genome editing technology. The CRISPR genome editing technology now used widely in mutant analysis in a variety of organisms is only a few years old, and the technology is still being optimized to enhance its specificity and efficiency. The central catalytic activity, RNA-guided cleavage that is directed against the genomic DNA, is carried out by the Cas9 endonuclease. Jennifer Doudna and colleagues use a fluorescence-based approach to define new features of Cas9 that control the specificity of RNA-guided DNA cleavage in CRISPR genome-editing technology. Cas9 is an RNA-guided DNA endonuclease that targets foreign DNA for destruction as part of a bacterial adaptive immune system mediated by clustered regularly interspaced short palindromic repeats (CRISPR)1,2. Together with single-guide RNAs3, Cas9 also functions as a powerful genome engineering tool in plants and animals4,5,6, and efforts are underway to increase the efficiency and specificity of DNA targeting for potential therapeutic applications7,8. Studies of off-target effects have shown that DNA binding is far more promiscuous than DNA cleavage9,10,11, yet the molecular cues that govern strand scission have not been elucidated. Here we show that the conformational state of the HNH nuclease domain directly controls DNA cleavage activity. Using intramolecular Förster resonance energy transfer experiments to detect relative orientations of the Cas9 catalytic domains when associated with on- and off-target DNA, we find that DNA cleavage efficiencies scale with the extent to which the HNH domain samples an activated conformation. We furthermore uncover a surprising mode of allosteric communication that ensures concerted firing of both Cas9 nuclease domains. Our results highlight a proofreading mechanism beyond initial protospacer adjacent motif (PAM) recognition12 and RNA–DNA base-pairing3 that serves as a final specificity checkpoint before DNA double-strand break formation.
- Top Cited PapersOpen Access
- 28 September 2014
- Published by Springer Nature in Nature
- Vol. 516 (7530) , 263-266
Programmable RNA recognition and cleavage by CRISPR/Cas9
In the presence of a short DNA oligonucleotide containing a protospacer adjacent motif, a guide-RNA-programmed Cas9 is able to specifically bind and/or cleave single-stranded RNA—this system can be used to isolate specific endogenous RNA transcripts from a cell lysate without any tag or modification. The bacterial CRISPR immune defence system, and its effector Cas9 in particular, have recently been exploited for sequence-specific genome editing in eukaryotic cells. Cas9 binds a guide RNA and in the presence of a DNA motif known as protospacer adjacent motif (PAM), is able to cleave the target DNA. New work by Jennifer Doudna and colleagues reveals the unexpected result that in the presence of a DNA oligomer containing PAM, a guide RNA-programmed Cas9 is able to cleave single-stranded RNA as well. They show that this system can also be used to isolate specific endogenous RNA transcripts, without any tag or modification, from a cell lysate. Thus, the system can be programmed to either bind or cut desired RNA targets, depending on the PAM used. This work and points the way towards possible new technologies for programmable RNA recognition. The CRISPR-associated protein Cas9 is an RNA-guided DNA endonuclease that uses RNA–DNA complementarity to identify target sites for sequence-specific double-stranded DNA (dsDNA) cleavage1,2,3,4,5. In its native context, Cas9 acts on DNA substrates exclusively because both binding and catalysis require recognition of a short DNA sequence, known as the protospacer adjacent motif (PAM), next to and on the strand opposite the twenty-nucleotide target site in dsDNA4,5,6,7. Cas9 has proven to be a versatile tool for genome engineering and gene regulation in a large range of prokaryotic and eukaryotic cell types, and in whole organisms8, but it has been thought to be incapable of targeting RNA5. Here we show that Cas9 binds with high affinity to single-stranded RNA (ssRNA) targets matching the Cas9-associated guide RNA sequence when the PAM is presented in trans as a separate DNA oligonucleotide. Furthermore, PAM-presenting oligonucleotides (PAMmers) stimulate site-specific endonucleolytic cleavage of ssRNA targets, similar to PAM-mediated stimulation of Cas9-catalysed DNA cleavage7. Using specially designed PAMmers, Cas9 can be specifically directed to bind or cut RNA targets while avoiding corresponding DNA sequences, and we demonstrate that this strategy enables the isolation of a specific endogenous messenger RNA from cells. These results reveal a fundamental connection between PAM binding and substrate selection by Cas9, and highlight the utility of Cas9 for programmable transcript recognition without the need for tags.
- 28 September 2014
- Published by Springer Nature in Nature
- Vol. 516 (7530) , 263-266
Top Cited PapersOpen AccessProgrammable RNA recognition and cleavage by CRISPR/Cas9
In the presence of a short DNA oligonucleotide containing a protospacer adjacent motif, a guide-RNA-programmed Cas9 is able to specifically bind and/or cleave single-stranded RNA—this system can be used to isolate specific endogenous RNA transcripts from a cell lysate without any tag or modification. The bacterial CRISPR immune defence system, and its effector Cas9 in particular, have recently been exploited for sequence-specific genome editing in eukaryotic cells. Cas9 binds a guide RNA and in the presence of a DNA motif known as protospacer adjacent motif (PAM), is able to cleave the target DNA. New work by Jennifer Doudna and colleagues reveals the unexpected result that in the presence of a DNA oligomer containing PAM, a guide RNA-programmed Cas9 is able to cleave single-stranded RNA as well. They show that this system can also be used to isolate specific endogenous RNA transcripts, without any tag or modification, from a cell lysate. Thus, the system can be programmed to either bind or cut desired RNA targets, depending on the PAM used. This work and points the way towards possible new technologies for programmable RNA recognition. The CRISPR-associated protein Cas9 is an RNA-guided DNA endonuclease that uses RNA–DNA complementarity to identify target sites for sequence-specific double-stranded DNA (dsDNA) cleavage1,2,3,4,5. In its native context, Cas9 acts on DNA substrates exclusively because both binding and catalysis require recognition of a short DNA sequence, known as the protospacer adjacent motif (PAM), next to and on the strand opposite the twenty-nucleotide target site in dsDNA4,5,6,7. Cas9 has proven to be a versatile tool for genome engineering and gene regulation in a large range of prokaryotic and eukaryotic cell types, and in whole organisms8, but it has been thought to be incapable of targeting RNA5. Here we show that Cas9 binds with high affinity to single-stranded RNA (ssRNA) targets matching the Cas9-associated guide RNA sequence when the PAM is presented in trans as a separate DNA oligonucleotide. Furthermore, PAM-presenting oligonucleotides (PAMmers) stimulate site-specific endonucleolytic cleavage of ssRNA targets, similar to PAM-mediated stimulation of Cas9-catalysed DNA cleavage7. Using specially designed PAMmers, Cas9 can be specifically directed to bind or cut RNA targets while avoiding corresponding DNA sequences, and we demonstrate that this strategy enables the isolation of a specific endogenous messenger RNA from cells. These results reveal a fundamental connection between PAM binding and substrate selection by Cas9, and highlight the utility of Cas9 for programmable transcript recognition without the need for tags.