lib310 Python package

Sample Usages

Protein Functional Annotation

# 1. import lib310 
import lib310

# 2. Get Spike SARS2 related proteins from database
seqs = lib310.db.fetch(
    name="SPIKE_SARS2",
    feature='sequence',
    limit=500
)

# 3. Instantiate a pre-trained GO Annotation machine learning model (e.g. TALE)
goa = lib310.ml.GoAnnotation.from_pretrained(model="prot_bert", v="latest")

# 4. Predict!
results = goa.run(seqs)

# 5. Visualization
lib310.plot.umap(results, color='protein_families')

Protein Generation

# 1. import lib310 
import lib310

# 2. Instantiate a pre-trained Generative Machine Learning model (e.g. GPT3)
lm = lib310.ml.AutoRegressiveLM.from_pretrained(model="ProtGPT3", v="latest")

# 3. Predict!
generated_sequences = lm.run(num_samples=1024)

# 4. Downstream Analysis...
clusters = lib310.tools.cluster(generated_sequences, method='kcluster')

# 5. Visualization
lib310.plot.umap(generated_sequences, clusters)