Making mouse predictions

AlphaGenome 2 makes predictions for mouse (Mus musculus) as well as human. This tutorial predicts RNA-seq coverage across the mouse alpha-globin cluster in erythroblasts.

Key activities:

• Load GENCODE vM23 (mm10) gene annotations

• List mouse reference conditions with supported_conditions

• Call predict_interval with organism=dna_client.Organism.MUS_MUSCULUS

Set up

# @title Install AlphaGenome

# @markdown Run this cell to install AlphaGenome.
from IPython.display import clear_output
! pip install alphagenome
clear_output()

# @title Imports
from alphagenome import colab_utils
from alphagenome.data import gene_annotation
from alphagenome.data import genome
from alphagenome.data import transcript as transcript_utils
from alphagenome.models.v2 import dna_client
from alphagenome.visualization import plot_components
import matplotlib.pyplot as plt
import pandas as pd

Store your API key in Colab Secrets for persistent access across sessions (see Installation), then initialize the client:

dna_model = dna_client.create(colab_utils.get_api_key())

Load GENCODE vM23 (mm10) gene annotations and build a TranscriptExtractor for the longest protein-coding transcript per gene (see GENCODE GTF format):

# The GTF file contains information on the location of all transcripts.
# Note that we use genome assembly mm10 for mice.
gtf = pd.read_feather(
    'https://storage.googleapis.com/alphagenome/reference/gencode/'
    'mm10/gencode.vM23.annotation.gtf.gz.feather'
)

# Get the longest transcript per gene
gtf_transcripts = gene_annotation.filter_protein_coding(gtf)
gtf_transcripts_longest = gene_annotation.filter_to_longest_transcript(
    gtf_transcripts
)

# Set up transcript extractors using the information in the GTF file.
transcript_extractor = transcript_utils.TranscriptExtractor(
    gtf_transcripts_longest
)

Make mouse predictions

Other tutorials use the default organism, human (Organism.HOMO_SAPIENS). For mouse predictions, pass organism=dna_client.Organism.MUS_MUSCULUS to both supported_conditions and predict_interval, and pass a mouse ontology CURIE via condition=. Expression conditioning (cell_conditioning=) is supported for human only:

dna_model.supported_conditions(organism=dna_client.Organism.MUS_MUSCULUS)

	condition	biosample_type	biosample_name	atac	cage	dnase	rna_seq	chip_histone	chip_tf	splice_sites	splice_site_usage	splice_junctions	contact_maps	sc_atac	sc_gene_expression	methylation
0	CL:0000038	primary_cell	erythroid progenitor cell	imputed_supported	imputed_exploratory	imputed_supported	imputed_supported	imputed_supported	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	observed	observed	imputed_supported
1	CL:0000048	primary_cell	multi fate stem cell	imputed_supported	imputed_exploratory	imputed_supported	imputed_supported	imputed_supported	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	observed	observed	imputed_supported
2	CL:0000136	primary_cell	adipocyte	imputed_supported	imputed_exploratory	observed	imputed_supported	imputed_supported	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_supported	imputed_supported	imputed_supported
3	CL:0000187	primary_cell	muscle cell	imputed_supported	imputed_exploratory	imputed_supported	observed	observed	observed	observed	observed	observed	imputed_exploratory	imputed_supported	imputed_supported	imputed_exploratory
4	CL:0000556	primary_cell	megakaryocyte	imputed_supported	imputed_exploratory	imputed_supported	observed	observed	imputed_exploratory	observed	observed	observed	imputed_exploratory	imputed_supported	imputed_supported	imputed_exploratory
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
715	UBERON:0008345	tissue	ileal epithelium	imputed_exploratory	observed	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory
716	UBERON:0008800	tissue	parietal endoderm	imputed_supported	imputed_exploratory	imputed_supported	imputed_supported	imputed_supported	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	observed	observed	imputed_supported
717	UBERON:0008945	tissue	extraembryonic endoderm	imputed_supported	imputed_exploratory	imputed_supported	imputed_supported	imputed_supported	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	observed	observed	imputed_supported
718	UBERON:0010230	tissue	eyeball of camera-type eye	imputed_exploratory	observed	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory
719	UBERON:0012314	tissue	embryonic facial prominence	observed	imputed_exploratory	observed	observed	observed	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_exploratory	imputed_supported	imputed_supported	imputed_supported

720 rows × 16 columns

Call predict_interval for a mouse genomic interval around the alpha-globin cluster (chr11:32206003-32337075) in mouse erythroblasts (CL:0000765; use dna_client.SEQUENCE_LENGTH_1MB for highest accuracy):

interval = genome.Interval.from_str('chr11:32206003-32337075').resize(
    dna_client.SEQUENCE_LENGTH_100KB
)

output = dna_model.predict_interval(
    interval=interval,
    organism=dna_client.Organism.MUS_MUSCULUS,
    requested_outputs=[dna_client.OutputType.RNA_SEQ],
    condition='CL:0000765',
)

output.rna_seq.values.shape

(131072, 12)

# Extract transcripts
transcripts = transcript_extractor.extract(interval)

# Plot
plot_components.plot(
    components=[
        # Add transcripts to plot
        plot_components.TranscriptAnnotation(transcripts),
        # Add RNA output
        plot_components.Tracks(
            output.rna_seq,
            ylabel_template=(
                '{biosample_name} ({strand})\n{name}\n{imputation_mode}'
            ),
        ),
    ],
    interval=interval,
    title='Predictions in mouse erythroblasts',
)

plt.show()