Raina W. Thomas
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Raina W. Thomas has a total of 89 co-authors. Here are the top co-authors by number of shared publications:
- Clare Bycroft
Google DeepMind, London, UK
3 shared publications - Guido Novati
Google DeepMind
3 shared publications - Natasha Latysheva
Google DeepMind, London, UK
3 shared publications - Pushmeet Kohli
Google DeepMind, London, UK
3 shared publications - Adam Gayoso
Google DeepMind; London, UK
2 shared publications - Alice L. Herneisen
Whitehead Institute for Biomedical Research and Department of Biology, Massachusetts Institute of Technology
2 shared publications
Latest Publications
- Open Access
AlphaGenome Atlas: in silico mutagenesis of the entire human genome improves prioritization and interpretation of non-coding variants
A major challenge in genomics is deciphering the functional consequences of non-coding genetic variation. Here we present AlphaGenome Atlas, a comprehensive resource that enables the joint interpretation and prioritization of variant effects across the entire human genome. Using AlphaGenome, we predicted the regulatory effects across thousands of molecular phenotypes for every possible human single nucleotide variant and many observed indels. These predictions were then used to derive a unified and interpretable AlphaGenome Variant Impact (AVI) score and to map cis-regulatory motifs across the genome. AVI achieved state-of-the-art performance across diverse benchmarks with improved prioritization of deleterious non-coding variants. Application of the combined Atlas resource helped solve an epileptic encephalopathy rare disease case, increased the statistical power to detect rare non-coding variants driving population-level phenotypes, and enhanced the mechanistic interpretation of these variants. Thus, AlphaGenome Atlas improves the prioritization and molecular interpretation of non-coding variants with genetic and clinical significance.
Open AccessAlphaGenome Atlas: in silico mutagenesis of the entire human genome improves prioritization and interpretation of non-coding variants
A major challenge in genomics is deciphering the functional consequences of non-coding genetic variation. Here we present AlphaGenome Atlas, a comprehensive resource that enables the joint interpretation and prioritization of variant effects across the entire human genome. Using AlphaGenome, we predicted the regulatory effects across thousands of molecular phenotypes for every possible human single nucleotide variant and many observed indels. These predictions were then used to derive a unified and interpretable AlphaGenome Variant Impact (AVI) score and to map cis-regulatory motifs across the genome. AVI achieved state-of-the-art performance across diverse benchmarks with improved prioritization of deleterious non-coding variants. Application of the combined Atlas resource helped solve an epileptic encephalopathy rare disease case, increased the statistical power to detect rare non-coding variants driving population-level phenotypes, and enhanced the mechanistic interpretation of these variants. Thus, AlphaGenome Atlas improves the prioritization and molecular interpretation of non-coding variants with genetic and clinical significance.
- Top Cited PapersOpen Access
- 28 January 2026
- Published by Springer Nature in Nature
- Vol. 649 (8099) , 1206-1218
Advancing regulatory variant effect prediction with AlphaGenome
Deep learning models that predict functional genomic measurements from DNA sequences are powerful tools for deciphering the genetic regulatory code. Existing methods involve a trade-off between input sequence length and prediction resolution, thereby limiting their modality scope and performance 1–5 . We present AlphaGenome, a unified DNA sequence model, which takes as input 1 Mb of DNA sequence and predicts thousands of functional genomic tracks up to single-base-pair resolution across diverse modalities. The modalities include gene expression, transcription initiation, chromatin accessibility, histone modifications, transcription factor binding, chromatin contact maps, splice site usage and splice junction coordinates and strength. Trained on human and mouse genomes, AlphaGenome matches or exceeds the strongest available external models in 25 of 26 evaluations of variant effect prediction. The ability of AlphaGenome to simultaneously score variant effects across all modalities accurately recapitulates the mechanisms of clinically relevant variants near the TAL1 oncogene 6 . To facilitate broader use, we provide tools for making genome track and variant effect predictions from sequence.
- 28 January 2026
- Published by Springer Nature in Nature
- Vol. 649 (8099) , 1206-1218
Top Cited PapersOpen AccessAdvancing regulatory variant effect prediction with AlphaGenome
Deep learning models that predict functional genomic measurements from DNA sequences are powerful tools for deciphering the genetic regulatory code. Existing methods involve a trade-off between input sequence length and prediction resolution, thereby limiting their modality scope and performance 1–5 . We present AlphaGenome, a unified DNA sequence model, which takes as input 1 Mb of DNA sequence and predicts thousands of functional genomic tracks up to single-base-pair resolution across diverse modalities. The modalities include gene expression, transcription initiation, chromatin accessibility, histone modifications, transcription factor binding, chromatin contact maps, splice site usage and splice junction coordinates and strength. Trained on human and mouse genomes, AlphaGenome matches or exceeds the strongest available external models in 25 of 26 evaluations of variant effect prediction. The ability of AlphaGenome to simultaneously score variant effects across all modalities accurately recapitulates the mechanisms of clinically relevant variants near the TAL1 oncogene 6 . To facilitate broader use, we provide tools for making genome track and variant effect predictions from sequence.
- Top Cited PapersOpen Access
AlphaGenome: advancing regulatory variant effect prediction with a unified DNA sequence model
Deep learning models that predict functional genomic measurements from DNA sequence are powerful tools for deciphering the genetic regulatory code. Existing methods trade off between input sequence length and prediction resolution, thereby limiting their modality scope and performance. We present AlphaGenome, which takes as input 1 megabase of DNA sequence and predicts thousands of functional genomic tracks up to single base pair resolution across diverse modalities – including gene expression, transcription initiation, chromatin accessibility, histone modifications, transcription factor binding, chro- matin contact maps, splice site usage, and splice junction coordinates and strength. Trained on human and mouse genomes, AlphaGenome matches or exceeds the strongest respective available external models on 24 out of 26 evaluations on variant effect prediction. AlphaGenome’s ability to simultaneously score variant effects across all modalities accurately recapitulates the mechanisms of clinically-relevant variants near the TAL1 oncogene. To facilitate broader use, we provide tools for making genome track and variant effect predictions from sequence.
Top Cited PapersOpen AccessAlphaGenome: advancing regulatory variant effect prediction with a unified DNA sequence model
Deep learning models that predict functional genomic measurements from DNA sequence are powerful tools for deciphering the genetic regulatory code. Existing methods trade off between input sequence length and prediction resolution, thereby limiting their modality scope and performance. We present AlphaGenome, which takes as input 1 megabase of DNA sequence and predicts thousands of functional genomic tracks up to single base pair resolution across diverse modalities – including gene expression, transcription initiation, chromatin accessibility, histone modifications, transcription factor binding, chro- matin contact maps, splice site usage, and splice junction coordinates and strength. Trained on human and mouse genomes, AlphaGenome matches or exceeds the strongest respective available external models on 24 out of 26 evaluations on variant effect prediction. AlphaGenome’s ability to simultaneously score variant effects across all modalities accurately recapitulates the mechanisms of clinically-relevant variants near the TAL1 oncogene. To facilitate broader use, we provide tools for making genome track and variant effect predictions from sequence.
- 8 July 2024
- Published by Springer Nature in Nature Microbiology
- Vol. 9 (9) , 2323-2343
CRISPR-based functional profiling of the Toxoplasma gondii genome during acute murine infection
Examining host–pathogen interactions in animals can capture aspects of infection that are obscured in cell culture. Using CRISPR-based screens, we functionally profile the entire genome of the apicomplexan parasite Toxoplasma gondii during murine infection. Barcoded gRNAs enabled bottleneck detection and mapping of population structures within parasite lineages. Over 300 genes with previously unknown roles in infection were found to modulate parasite fitness in mice. Candidates span multiple axes of host–parasite interaction. Rhoptry Apical Surface Protein 1 was characterized as a mediator of host-cell tropism that facilitates repeated invasion attempts. GTP cyclohydrolase I was also required for fitness in mice and druggable through a repurposed compound, 2,4-diamino-6-hydroxypyrimidine. This compound synergized with pyrimethamine against T. gondii and malaria-causing Plasmodium falciparum parasites. This work represents a complete survey of an apicomplexan genome during infection of an animal host and points to novel interfaces of host–parasite interaction.
- 8 July 2024
- Published by Springer Nature in Nature Microbiology
- Vol. 9 (9) , 2323-2343
CRISPR-based functional profiling of the Toxoplasma gondii genome during acute murine infection
Examining host–pathogen interactions in animals can capture aspects of infection that are obscured in cell culture. Using CRISPR-based screens, we functionally profile the entire genome of the apicomplexan parasite Toxoplasma gondii during murine infection. Barcoded gRNAs enabled bottleneck detection and mapping of population structures within parasite lineages. Over 300 genes with previously unknown roles in infection were found to modulate parasite fitness in mice. Candidates span multiple axes of host–parasite interaction. Rhoptry Apical Surface Protein 1 was characterized as a mediator of host-cell tropism that facilitates repeated invasion attempts. GTP cyclohydrolase I was also required for fitness in mice and druggable through a repurposed compound, 2,4-diamino-6-hydroxypyrimidine. This compound synergized with pyrimethamine against T. gondii and malaria-causing Plasmodium falciparum parasites. This work represents a complete survey of an apicomplexan genome during infection of an animal host and points to novel interfaces of host–parasite interaction.
- Open Access
Functional profiling of the Toxoplasma genome during acute mouse infection
SUMMARY: Within a host, pathogens encounter a diverse and changing landscape of cell types, nutrients, and immune responses. Examining host-pathogen interactions in animal models can therefore reveal aspects of infection absent from cell culture. We use CRISPR-based screens to functionally profile the entire genome of the model apicomplexan parasite Toxoplasma gondii during mouse infection. Barcoded gRNAs were used to track mutant parasite lineages, enabling detection of bottlenecks and mapping of population structures. We uncovered over 300 genes that modulate parasite fitness in mice with previously unknown roles in infection. These candidates span multiple axes of host-parasite interaction, including determinants of tropism, host organelle remodeling, and metabolic rewiring. We mechanistically characterized three novel candidates, including GTP cyclohydrolase I, against which a small-molecule inhibitor could be repurposed as an antiparasitic compound. This compound exhibited antiparasitic activity against T. gondii and Plasmodium falciparum, the most lethal agent of malaria. Taken together, we present the first complete survey of an apicomplexan genome during infection of an animal host, and point to novel interfaces of host-parasite interaction that may offer new avenues for treatment.
Open AccessFunctional profiling of the Toxoplasma genome during acute mouse infection
SUMMARY: Within a host, pathogens encounter a diverse and changing landscape of cell types, nutrients, and immune responses. Examining host-pathogen interactions in animal models can therefore reveal aspects of infection absent from cell culture. We use CRISPR-based screens to functionally profile the entire genome of the model apicomplexan parasite Toxoplasma gondii during mouse infection. Barcoded gRNAs were used to track mutant parasite lineages, enabling detection of bottlenecks and mapping of population structures. We uncovered over 300 genes that modulate parasite fitness in mice with previously unknown roles in infection. These candidates span multiple axes of host-parasite interaction, including determinants of tropism, host organelle remodeling, and metabolic rewiring. We mechanistically characterized three novel candidates, including GTP cyclohydrolase I, against which a small-molecule inhibitor could be repurposed as an antiparasitic compound. This compound exhibited antiparasitic activity against T. gondii and Plasmodium falciparum, the most lethal agent of malaria. Taken together, we present the first complete survey of an apicomplexan genome during infection of an animal host, and point to novel interfaces of host-parasite interaction that may offer new avenues for treatment.