Standardize metadata on-the-fly

This use cases runs on a LaminDB instance with populated CellType and Pathway registries. Make sure you run the GO Ontology notebook before executing this use case.

Here, we demonstrate how to standardize the metadata on-the-fly during cell type annotation and pathway enrichment analysis using these two registries.

For more information, see:

!lamin load use-cases-registries
Entity has to be a laminhub URL or 'artifact' or 'transform'
import lamindb as ln
import bionty as bt
from lamin_usecases import datasets as ds
import scanpy as sc
import matplotlib.pyplot as plt
import celltypist
import gseapy as gp
 connected lamindb: testuser1/use-cases-registries
sc.settings.set_figure_params(dpi=50, facecolor="white")
ln.track("hsPU1OENv0LS0000")
 created Transform('hsPU1OEN'), started new Run('bg8vT3Xj') at 2024-12-20 15:05:36 UTC
 notebook imports: bionty==0.53.2 celltypist==1.6.3 gseapy==1.1.4 lamin_usecases==0.0.1 lamindb==0.77.3 matplotlib==3.10.0 scanpy==1.10.4

An interferon-beta treated dataset

A small peripheral blood mononuclear cell dataset that is split into control and stimulated groups. The stimulated group was treated with interferon beta.

Let’s load the dataset and perform some preprocessing:

adata = ds.anndata_seurat_ifnb(preprocess=False, populate_registries=True)
adata

AnnData object with n_obs × n_vars = 13999 × 9943
    obs: 'stim'
    var: 'symbol'
sc.pp.normalize_total(adata, target_sum=1e4)
sc.pp.log1p(adata)
sc.pp.highly_variable_genes(adata, n_top_genes=2000)
sc.pp.pca(adata, n_comps=20)
sc.pp.neighbors(adata, n_pcs=10)
sc.tl.umap(adata)

Analysis: cell type annotation using CellTypist

model = celltypist.models.Model.load(model="Immune_All_Low.pkl")
Hide code cell output
🔎 No available models. Downloading...
📜 Retrieving model list from server https://celltypist.cog.sanger.ac.uk/models/models.json
📚 Total models in list: 53
📂 Storing models in /home/runner/.celltypist/data/models
💾 Downloading model [1/53]: Immune_All_Low.pkl
💾 Downloading model [2/53]: Immune_All_High.pkl
💾 Downloading model [3/53]: Adult_COVID19_PBMC.pkl
💾 Downloading model [4/53]: Adult_CynomolgusMacaque_Hippocampus.pkl
💾 Downloading model [5/53]: Adult_Human_MTG.pkl
💾 Downloading model [6/53]: Adult_Human_PancreaticIslet.pkl
💾 Downloading model [7/53]: Adult_Human_PrefrontalCortex.pkl
💾 Downloading model [8/53]: Adult_Human_Skin.pkl
💾 Downloading model [9/53]: Adult_Human_Vascular.pkl
💾 Downloading model [10/53]: Adult_Mouse_Gut.pkl
💾 Downloading model [11/53]: Adult_Mouse_OlfactoryBulb.pkl
💾 Downloading model [12/53]: Adult_Pig_Hippocampus.pkl
💾 Downloading model [13/53]: Adult_RhesusMacaque_Hippocampus.pkl
💾 Downloading model [14/53]: Autopsy_COVID19_Lung.pkl
💾 Downloading model [15/53]: COVID19_HumanChallenge_Blood.pkl
💾 Downloading model [16/53]: COVID19_Immune_Landscape.pkl
💾 Downloading model [17/53]: Cells_Adult_Breast.pkl
💾 Downloading model [18/53]: Cells_Fetal_Lung.pkl
💾 Downloading model [19/53]: Cells_Human_Tonsil.pkl
💾 Downloading model [20/53]: Cells_Intestinal_Tract.pkl
💾 Downloading model [21/53]: Cells_Lung_Airway.pkl
💾 Downloading model [22/53]: Developing_Human_Brain.pkl
💾 Downloading model [23/53]: Developing_Human_Gonads.pkl
💾 Downloading model [24/53]: Developing_Human_Hippocampus.pkl
💾 Downloading model [25/53]: Developing_Human_Organs.pkl
💾 Downloading model [26/53]: Developing_Human_Thymus.pkl
💾 Downloading model [27/53]: Developing_Mouse_Brain.pkl
💾 Downloading model [28/53]: Developing_Mouse_Hippocampus.pkl
💾 Downloading model [29/53]: Fetal_Human_AdrenalGlands.pkl
💾 Downloading model [30/53]: Fetal_Human_Pancreas.pkl
💾 Downloading model [31/53]: Fetal_Human_Pituitary.pkl
💾 Downloading model [32/53]: Fetal_Human_Retina.pkl
💾 Downloading model [33/53]: Fetal_Human_Skin.pkl
💾 Downloading model [34/53]: Healthy_Adult_Heart.pkl
💾 Downloading model [35/53]: Healthy_COVID19_PBMC.pkl
💾 Downloading model [36/53]: Healthy_Human_Liver.pkl
💾 Downloading model [37/53]: Healthy_Mouse_Liver.pkl
💾 Downloading model [38/53]: Human_AdultAged_Hippocampus.pkl
💾 Downloading model [39/53]: Human_Colorectal_Cancer.pkl
💾 Downloading model [40/53]: Human_Developmental_Retina.pkl
💾 Downloading model [41/53]: Human_Embryonic_YolkSac.pkl
💾 Downloading model [42/53]: Human_IPF_Lung.pkl
💾 Downloading model [43/53]: Human_Longitudinal_Hippocampus.pkl
💾 Downloading model [44/53]: Human_Lung_Atlas.pkl
💾 Downloading model [45/53]: Human_PF_Lung.pkl
💾 Downloading model [46/53]: Human_Placenta_Decidua.pkl
💾 Downloading model [47/53]: Lethal_COVID19_Lung.pkl
💾 Downloading model [48/53]: Mouse_Dentate_Gyrus.pkl
💾 Downloading model [49/53]: Mouse_Isocortex_Hippocampus.pkl
💾 Downloading model [50/53]: Mouse_Postnatal_DentateGyrus.pkl
💾 Downloading model [51/53]: Mouse_Whole_Brain.pkl
💾 Downloading model [52/53]: Nuclei_Lung_Airway.pkl
💾 Downloading model [53/53]: Pan_Fetal_Human.pkl
predictions = celltypist.annotate(
    adata, model="Immune_All_Low.pkl", majority_voting=True
)
adata.obs["cell_type_celltypist"] = predictions.predicted_labels.majority_voting
🔬 Input data has 13999 cells and 9943 genes
🔗 Matching reference genes in the model
🧬 3699 features used for prediction
⚖️ Scaling input data
🖋️ Predicting labels
✅ Prediction done!
👀 Detected a neighborhood graph in the input object, will run over-clustering on the basis of it
⛓️ Over-clustering input data with resolution set to 10
🗳️ Majority voting the predictions
✅ Majority voting done!
adata.obs["cell_type_celltypist"] = bt.CellType.standardize(
    adata.obs["cell_type_celltypist"]
)
sc.pl.umap(
    adata,
    color=["cell_type_celltypist", "stim"],
    frameon=False,
    legend_fontsize=10,
    wspace=0.4,
)
... storing 'cell_type_celltypist' as categorical
_images/54bf9e5faa046818a0891d3030549bdbb49344c5ab6c935b0f5daa321ed79df4.png

Analysis: Pathway enrichment analysis using Enrichr

This analysis is based on the GSEApy scRNA-seq Example.

First, we compute differentially expressed genes using a Wilcoxon test between stimulated and control cells.

# compute differentially expressed genes
sc.tl.rank_genes_groups(
    adata,
    groupby="stim",
    use_raw=False,
    method="wilcoxon",
    groups=["STIM"],
    reference="CTRL",
)

rank_genes_groups_df = sc.get.rank_genes_groups_df(adata, "STIM")
rank_genes_groups_df.head()
names scores logfoldchanges pvals pvals_adj
0 ISG15 99.456627 7.132626 0.0 0.0
1 ISG20 96.736649 5.074122 0.0 0.0
2 IFI6 94.972954 5.828700 0.0 0.0
3 IFIT3 92.482460 7.432328 0.0 0.0
4 IFIT1 90.699036 8.053429 0.0 0.0

Next, we filter out up/down-regulated differentially expressed gene sets:

degs_up = rank_genes_groups_df[
    (rank_genes_groups_df["logfoldchanges"] > 0)
    & (rank_genes_groups_df["pvals_adj"] < 0.05)
]
degs_dw = rank_genes_groups_df[
    (rank_genes_groups_df["logfoldchanges"] < 0)
    & (rank_genes_groups_df["pvals_adj"] < 0.05)
]

degs_up.shape, degs_dw.shape
((541, 5), (937, 5))

Run pathway enrichment analysis on DEGs and plot top 10 pathways:

enr_up = gp.enrichr(degs_up.names, gene_sets="GO_Biological_Process_2023").res2d
gp.dotplot(enr_up, figsize=(2, 3), title="Up", cmap=plt.cm.autumn_r);
enr_dw = gp.enrichr(degs_dw.names, gene_sets="GO_Biological_Process_2023").res2d
gp.dotplot(enr_dw, figsize=(2, 3), title="Down", cmap=plt.cm.winter_r);

Annotate & save dataset

gRegister new features and labels (check out more details here):

new_features = ln.Feature.from_df(adata.obs)
ln.save(new_features)
new_labels = [ln.ULabel(name=i) for i in adata.obs["stim"].unique()]
ln.save(new_labels)
! You have few permissible values for feature stim, consider dtype 'cat' instead of 'str'
! You have few permissible values for feature cell_type_celltypist, consider dtype 'cat' instead of 'str'
features = ln.Feature.lookup()

Register dataset using a Artifact object:

artifact = ln.Artifact.from_anndata(
    adata,
    description="seurat_ifnb_activated_Bcells",
).save()
# TODO: rewrite based on ln.Curator.from_anndata()
# artifact.features._add_set_from_anndata(
#     var_field=bt.Gene.symbol,
#     organism="human",  # optionally, globally set organism via bt.settings.organism = "human"
# )
# cell_type_records = bt.CellType.from_values(adata.obs["cell_type_celltypist"])
# artifact.labels.add(cell_type_records, features.cell_type_celltypist)
# stim_records = ln.ULabel.from_values(adata.obs["stim"])
# artifact.labels.add(stim_records, features.stim)

Querying pathways

Querying for pathways contains “interferon-beta” in the name:

bt.Pathway.filter(name__contains="interferon-beta").df()
uid name ontology_id abbr synonyms description source_id run_id created_at created_by_id
id
684 1l4z0v8W cellular response to interferon-beta GO:0035458 None cellular response to fiblaferon|cellular respo... Any Process That Results In A Change In State ... 86 None 2024-12-20 15:04:53.293956+00:00 1
2130 1NzHDJDi negative regulation of interferon-beta production GO:0032688 None down regulation of interferon-beta production|... Any Process That Stops, Prevents, Or Reduces T... 86 None 2024-12-20 15:04:53.352050+00:00 1
3127 3x0xmK1y positive regulation of interferon-beta production GO:0032728 None up-regulation of interferon-beta production|up... Any Process That Activates Or Increases The Fr... 86 None 2024-12-20 15:04:53.403539+00:00 1
4334 54R2a0el regulation of interferon-beta production GO:0032648 None regulation of IFN-beta production Any Process That Modulates The Frequency, Rate... 86 None 2024-12-20 15:04:53.456787+00:00 1
4953 3VZq4dMe response to interferon-beta GO:0035456 None response to fiblaferon|response to fibroblast ... Any Process That Results In A Change In State ... 86 None 2024-12-20 15:04:53.484359+00:00 1

Query pathways from a gene:

bt.Pathway.filter(genes__symbol="KIR2DL1").df()
uid name ontology_id abbr synonyms description source_id run_id created_at created_by_id
id
1346 7S7qlEkG immune response-inhibiting cell surface recept... GO:0002767 None immune response-inhibiting cell surface recept... The Series Of Molecular Signals Initiated By A... 86 None 2024-12-20 15:04:53.319836+00:00 1

Query artifacts from a pathway:

ln.Artifact.filter(feature_sets__pathways__name__icontains="interferon-beta").first()
Artifact(uid='umqzYDd43eJeS3zf0000', is_latest=True, description='seurat_ifnb_activated_Bcells', suffix='.h5ad', type='dataset', size=215085769, hash='J9oa_JXX4wZ-frPnlk2slL', _hash_type='sha1-fl', _accessor='AnnData', visibility=1, _key_is_virtual=True, storage_id=1, transform_id=1, run_id=1, created_by_id=1, created_at=2024-12-20 15:09:05 UTC)

Query featuresets from a pathway to learn from which geneset this pathway was computed:

pathway = bt.Pathway.get(ontology_id="GO:0035456")
pathway
Pathway(uid='3VZq4dMe', name='response to interferon-beta', ontology_id='GO:0035456', synonyms='response to fiblaferon|response to fibroblast interferon|response to interferon beta', description='Any Process That Results In A Change In State Or Activity Of A Cell Or An Organism (In Terms Of Movement, Secretion, Enzyme Production, Gene Expression, Etc.) As A Result Of An Interferon-Beta Stimulus. Interferon-Beta Is A Type I Interferon.', created_by_id=1, source_id=86, created_at=2024-12-20 15:04:53 UTC)
degs = ln.FeatureSet.get(pathways__ontology_id=pathway.ontology_id)

Now we can get the list of genes that are differentially expressed and belong to this pathway:

contributing_genes = pathway.genes.all() & degs.genes.all()
contributing_genes.list("symbol")
['PLSCR1',
 'IFITM2',
 'IFITM1',
 'XAF1',
 'SHFL',
 'MNDA',
 'OAS1',
 'IFI16',
 'STAT1',
 'AIM2',
 'PNPT1',
 'IFITM3',
 'CALM1',
 'BST2',
 'IRF1']
# clean up test instance
!lamin delete --force use-cases-registries
!rm -r ./use-cases-registries
Hide code cell output
Traceback (most recent call last):
  File "/opt/hostedtoolcache/Python/3.11.11/x64/bin/lamin", line 8, in <module>
    sys.exit(main())
             ^^^^^^
  File "/opt/hostedtoolcache/Python/3.11.11/x64/lib/python3.11/site-packages/rich_click/rich_command.py", line 367, in __call__
    return super().__call__(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/hostedtoolcache/Python/3.11.11/x64/lib/python3.11/site-packages/click/core.py", line 1157, in __call__
    return self.main(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/hostedtoolcache/Python/3.11.11/x64/lib/python3.11/site-packages/rich_click/rich_command.py", line 152, in main
    rv = self.invoke(ctx)
         ^^^^^^^^^^^^^^^^
  File "/opt/hostedtoolcache/Python/3.11.11/x64/lib/python3.11/site-packages/click/core.py", line 1688, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/hostedtoolcache/Python/3.11.11/x64/lib/python3.11/site-packages/click/core.py", line 1434, in invoke
    return ctx.invoke(self.callback, **ctx.params)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/hostedtoolcache/Python/3.11.11/x64/lib/python3.11/site-packages/click/core.py", line 783, in invoke
    return __callback(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/hostedtoolcache/Python/3.11.11/x64/lib/python3.11/site-packages/lamin_cli/__main__.py", line 209, in delete
    return delete(instance, force=force)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/hostedtoolcache/Python/3.11.11/x64/lib/python3.11/site-packages/lamindb_setup/_delete.py", line 102, in delete
    n_objects = check_storage_is_empty(
                ^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/hostedtoolcache/Python/3.11.11/x64/lib/python3.11/site-packages/lamindb_setup/core/upath.py", line 836, in check_storage_is_empty
    raise InstanceNotEmpty(message)
lamindb_setup.core.upath.InstanceNotEmpty: Storage '/home/runner/work/lamin-usecases/lamin-usecases/docs/use-cases-registries/.lamindb' contains 1 objects - delete them prior to deleting the instance