02: Expression plot
Create an RNA isoform structure plot with an RNA isoform expression plot beside it
[7]:
import RNApysoforms as RNApy
[8]:
## Path to your ENSEMBL GTF file and counts matrix file
ensembl_gtf_path = "../dash_apps/RNApysoforms/tests/test_data/Homo_sapiens_chr21_and_Y.GRCh38.110.gtf"
counts_matrix_path = "../dash_apps/RNApysoforms/tests/test_data/counts_matrix_chr21_and_Y.tsv"
[9]:
## Read ENSEMBL gtf and counts matrix
annotation = RNApy.read_ensembl_gtf(ensembl_gtf_path)
counts_matrix = RNApy.read_expression_matrix(expression_matrix_path=counts_matrix_path)
[10]:
## Filter gene name in annotation and counts matrix.
sod1_annotation, sod1_counts_matrix = RNApy.gene_filtering(annotation=annotation, expression_matrix=counts_matrix, target_gene="SOD1")
sod1_counts_matrix.head()
[10]:
shape: (5, 4)
| transcript_id | gene_id | sample_id | counts |
|---|---|---|---|
| str | str | str | f64 |
| "ENST00000476106" | "ENSG00000142168" | "sample_1" | 0.0 |
| "ENST00000476106" | "ENSG00000142168" | "sample_4" | 0.0 |
| "ENST00000476106" | "ENSG00000142168" | "sample_7" | 0.0 |
| "ENST00000476106" | "ENSG00000142168" | "sample_2" | 0.0 |
| "ENST00000476106" | "ENSG00000142168" | "sample_6" | 0.0 |
[11]:
"""
Rescale introns (no need to run function "to_intron", shorten_gaps() already does this
by default if introns aren't already included in annotation.
"""
sod1_annotation = RNApy.shorten_gaps(sod1_annotation)
sod1_annotation.head()
[11]:
shape: (5, 13)
| gene_id | gene_name | transcript_id | transcript_name | transcript_biotype | seqnames | strand | type | start | end | exon_number | rescaled_start | rescaled_end |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| str | str | str | str | str | str | str | str | i64 | i64 | i64 | i64 | i64 |
| "ENSG00000142168" | "SOD1" | "ENST00000270142" | "SOD1-201" | "protein_coding" | "21" | "+" | "exon" | 31659693 | 31659841 | 1 | 29 | 177 |
| "ENSG00000142168" | "SOD1" | "ENST00000270142" | "SOD1-201" | "protein_coding" | "21" | "+" | "CDS" | 31659770 | 31659841 | 1 | 106 | 177 |
| "ENSG00000142168" | "SOD1" | "ENST00000270142" | "SOD1-201" | "protein_coding" | "21" | "+" | "intron" | 31659841 | 31663790 | 1 | 177 | 1431 |
| "ENSG00000142168" | "SOD1" | "ENST00000270142" | "SOD1-201" | "protein_coding" | "21" | "+" | "CDS" | 31663790 | 31663886 | 2 | 1431 | 1527 |
| "ENSG00000142168" | "SOD1" | "ENST00000270142" | "SOD1-201" | "protein_coding" | "21" | "+" | "exon" | 31663790 | 31663886 | 2 | 1431 | 1527 |
[12]:
"""
Create traces for plotting, the expression plot will come out in
the order of columns passed on the `expression_columns` parameters.
This is important if you are passing multiple expression columns
like CPM and relative abundance.
"""
traces = RNApy.make_traces(annotation=sod1_annotation, expression_matrix=sod1_counts_matrix,
x_start="rescaled_start", x_end="rescaled_end",
y='transcript_id', annotation_hue="transcript_biotype",
hover_start="start", hover_end="end",
expression_columns=["counts"])
"""
Put traces into the figure. The order of `subplot_titles` is important.
The first plot will always be "Transcript Structure" if you passed an annotation
to make the traces. After that the order of the expression plots is determined
by the `expression_columns` parameter passed to the `make_traces()` function.
"""
fig = RNApy.make_plot(traces = traces, subplot_titles = ["Transcript Structure", "Counts"], width=1200, height=500)
## Show figure
fig.show()