Top Tweets for #EnzymeAnnotation
An integrated AI knowledge graph framework of bacterial enzymology and metabolism @PNASNews
1. This study introduces IBIS, a unified framework that integrates deep learning and knowledge graphs to annotate and compare bacterial metabolism at scale, offering comprehensive insight into both primary and specialized metabolism.
2. IBIS-Enzyme, a Transformer model, encodes protein sequences into embeddings, enabling accurate prediction of enzyme classes (F1 = 0.95), outperforming tools like DeepEC and CLEAN and generalizing to untrained EC numbers.
3. The framework enables domain-level annotation with IBIS-Domain, which outperforms prior tools in predicting substrate specificity of biosynthetic enzymes such as NRPS and PKS.
4. IBIS-SM, a Graphormer model, identifies biosynthetic gene clusters (BGCs) without relying on pHMMs, showing superior accuracy in BGC boundary calling and chemotype classification across 36 classes.
5. IBIS-BGC generates vector embeddings for entire BGCs, enabling rapid, structure-aware clustering of nearly 2 million BGCs into 219,587 families in under one hour, a massive leap over current methods like BiG-SLiCE.
6. Embedding-based classification enables IBIS to annotate proteins and domains 5–20x faster than traditional methods like DIAMOND or PFAM hmmscan, enabling high-throughput genome annotation.
7. IBIS-KG connects enzyme annotations with ecological and taxonomic metadata, allowing discovery of niche-specific specialized metabolites and uncharacterized pathways, especially in underexplored bacterial taxa.
8. Experimental validation confirmed the utility of IBIS predictions by isolating novel metabolite derivatives, such as new analogs of xenoamicin and amicoumacin, from previously uncharacterized bacteria.
9. IBIS-KG reveals that 6–25% of enzymes across various bacteria are unassigned to known metabolism, highlighting vast areas of uncharacterized enzymatic diversity for further exploration.
10. The framework identifies over 107,000 genomic regions enriched with uncharacterized enzymes, many statistically associated with specific ecological niches, suggesting new hypotheses for bacterial functionality.
11. IBIS offers a scalable, flexible alternative to annotation pipelines like antiSMASH and PRISM, combining generalizable learning with interpretable, context-aware metabolic inference.
12. By enabling vector-based lookup of annotated embeddings, IBIS facilitates scalable and future-proof bacterial metabolism discovery as genomic data continues to expand.
💻Code: github.com/Magarvey-Lab/IBIS
📜Paper: pnas.org/doi/10.1073/pnas.24…
#Microbiome #Metabolomics #BacterialGenomics #Bioinformatics #DeepLearning #GraphNeuralNetworks #KnowledgeGraph #MetabolicEngineering #DrugDiscovery #EnzymeAnnotation

Most Popular Users

Elon Musk 
@elonmusk
240.3M followers

Barack Obama 
@barackobama
118.6M followers

Cristiano Ronaldo 
@cristiano
114.4M followers

Donald J. Trump 
@realdonaldtrump
111.8M followers

Narendra Modi 
@narendramodi
106.9M followers

Rihanna 
@rihanna
98.6M followers

NASA 
@nasa
92.2M followers

Justin Bieber 
@justinbieber
91.7M followers

KATY PERRY 
@katyperry
89.9M followers

Taylor Swift 
@taylorswift13
83.9M followers

Lady Gaga 
@ladygaga
75.4M followers

Virat Kohli 
@imvkohli
73.3M followers

Kim Kardashian 
@kimkardashian
70.8M followers

YouTube 
@youtube
68.6M followers

Neymar Jr 
@neymarjr
66.3M followers

Bill Gates 
@billgates
65.2M followers

Selena Gomez 
@selenagomez
63M followers

The Ellen Show
@theellenshow
62.1M followers

CNN 
@cnn
61.7M followers

X 
@x
60.6M followers
