Project flow

LaminDB allows tracking data lineage on the entire project level.

Here, we walk through exemplified app uploads, pipelines & notebooks following Schmidt et al., 2022.

A CRISPR screen reading out a phenotypic endpoint on T cells is paired with scRNA-seq to generate insights into IFN-γ production.

These insights get linked back to the original data through the steps taken in the project to provide context for interpretation & future decision making.

More specifically: Why should I care about data flow?

Data flow tracks data sources & transformations to trace biological insights, verify experimental outcomes, meet regulatory standards, increase the robustness of research and optimize the feedback loop of team-wide learning iterations.

While tracking data flow is easier when it’s governed by deterministic pipelines, it becomes hard when it’s governed by interactive human-driven analyses.

LaminDB interfaces workflow mangers for the former and embraces the latter.

Setup

Init a test instance:

!lamin init --storage ./mydata

Show code cell output Hide code cell output

💡 connected lamindb: testuser1/mydata

Import lamindb:

import lamindb as ln
from IPython.display import Image, display

💡 connected lamindb: testuser1/mydata

Steps

In the following, we walk through exemplified steps covering different types of transforms (Transform).

Note

The full notebooks are in this repository.

App upload of phenotypic data

Register data through app upload from wetlab by testuser1:

# This function mimics the upload of artifacts via the UI
# In reality, you simply drag and drop files into the UI
def mock_upload_crispra_result_app():
    ln.setup.login("testuser1")
    transform = ln.Transform(name="Upload GWS CRISPRa result", type="upload")
    ln.track(transform=transform)
    output_path = ln.core.datasets.schmidt22_crispra_gws_IFNG(ln.settings.storage)
    output_file = ln.Artifact(
        output_path, description="Raw data of schmidt22 crispra GWS"
    )
    output_file.save()

mock_upload_crispra_result_app()

Show code cell output Hide code cell output

💡 saved: Transform(uid='IklsxKrGLRImLZh0', name='Upload GWS CRISPRa result', type='upload', updated_at=2024-04-22 10:27:14 UTC, created_by_id=1)

💡 saved: Run(uid='EWfQAIrW3wbA1uOGeERV', transform_id=1, created_by_id=1)

Hit identification in notebook

Access, transform & register data in drylab by testuser2:

# This function mimics the hit identification notebook
# In reality, you would run this in a notebook titled "GWS CRIPSRa analysis"
def mock_hit_identification_notebook():
    # log in as another user
    ln.setup.login("testuser2")

    # create a new transform to mimic a new notebook
    # (in reality you just run ln.track() in a notebook and you don't have to manage runs)
    transform = ln.Transform(name="GWS CRIPSRa analysis", type="notebook")
    transform.save()
    run = ln.Run(transform=transform)
    run.save()

    # access the upload artifact
    input_file = ln.Artifact.filter(key="schmidt22-crispra-gws-IFNG.csv").one()

    # identify hits
    input_df = input_file.load(is_run_input=run).set_index("id")
    output_df = input_df[input_df["pos|fdr"] < 0.01].copy()

    # register hits in output artifact
    ln.Artifact.from_df(output_df, description="hits from schmidt22 crispra GWS", run=run).save()

mock_hit_identification_notebook()

Inspect data flow:

artifact = ln.Artifact.filter(description="hits from schmidt22 crispra GWS").one()
artifact.view_lineage()

Sequencer upload

Upload files from sequencer via script chromium_10x_upload.py:

!python project-flow-scripts/chromium_10x_upload.py

💡 connected lamindb: testuser1/mydata

💡 saved: Transform(uid='qCJPkOuZAi9q5zKv', name='chromium_10x_upload.py', key='chromium_10x_upload.py', version='1', type='script', updated_at=2024-04-22 10:27:17 UTC, created_by_id=1)
💡 saved: Run(uid='BykoALMl9mp8dzfXUNqm', transform_id=3, created_by_id=1)

✅ saved transform.source_code: Artifact(uid='zKbQLfb8wAhavsv2ubkE', suffix='.py', description='Source of transform qCJPkOuZAi9q5zKv', version='1', size=474, hash='o-QoKgEZGxbk5oBtcAKoWw', hash_type='md5', visibility=0, key_is_virtual=True, updated_at=2024-04-22 10:27:18 UTC, storage_id=1, created_by_id=1)
✅ saved run.environment: Artifact(uid='3McctApJlP650n8VE3cY', suffix='.txt', description='requirements.txt', size=3400, hash='lrHLv26VOiHp8JkBy2B7zw', hash_type='md5', visibility=0, key_is_virtual=True, updated_at=2024-04-22 10:27:18 UTC, storage_id=1, created_by_id=1)

scRNA-seq bioinformatics pipeline

Process uploaded files using a script or workflow manager: Pipelines and obtain 3 output files in a directory filtered_feature_bc_matrix/:

cellranger.py

!python project-flow-scripts/cellranger.py

💡 connected lamindb: testuser1/mydata

💡 saved: Transform(uid='w6eKweXO1stJDZiW', name='Cell Ranger', version='7.2.0', type='pipeline', reference='https://www.10xgenomics.com/support/software/cell-ranger/7.2', updated_at=2024-04-22 10:27:20 UTC, created_by_id=2)
💡 saved: Run(uid='oCsKHG1XuH1jK4o0J1BX', transform_id=4, created_by_id=2)

❗ this creates one artifact per file in the directory - you might simply call ln.Artifact(dir) to get one artifact for the entire directory

postprocess_cellranger.py

!python project-flow-scripts/postprocess_cellranger.py

💡 connected lamindb: testuser1/mydata

💡 saved: Transform(uid='YqmbO6oMXjRj65cN', name='postprocess_cellranger.py', key='postprocess_cellranger.py', version='2', type='script', updated_at=2024-04-22 10:27:22 UTC, created_by_id=2)
💡 saved: Run(uid='nCDH4kLNiN2ZhCjuwVuO', transform_id=5, created_by_id=2)

✅ saved transform.source_code: Artifact(uid='j9i7StnK6LjbBRQ7fpN3', suffix='.py', description='Source of transform YqmbO6oMXjRj65cN', version='2', size=495, hash='iLSbWXZ-j7pkIgzO0i6c0w', hash_type='md5', visibility=0, key_is_virtual=True, updated_at=2024-04-22 10:27:23 UTC, storage_id=1, created_by_id=2)
❗ returning existing artifact with same hash: Artifact(uid='3McctApJlP650n8VE3cY', suffix='.txt', description='requirements.txt', size=3400, hash='lrHLv26VOiHp8JkBy2B7zw', hash_type='md5', visibility=0, key_is_virtual=True, updated_at=2024-04-22 10:27:18 UTC, storage_id=1, created_by_id=1)
✅ saved run.environment: Artifact(uid='3McctApJlP650n8VE3cY', suffix='.txt', description='requirements.txt', size=3400, hash='lrHLv26VOiHp8JkBy2B7zw', hash_type='md5', visibility=0, key_is_virtual=True, updated_at=2024-04-22 10:27:18 UTC, storage_id=1, created_by_id=1)

Inspect data flow:

output_file = ln.Artifact.filter(description="perturbseq counts").one()
output_file.view_lineage()

Integrate scRNA-seq & phenotypic data

Integrate data in a notebook:

# This function mimics the integrated analysis notebook
# In reality, you would run this in a notebook titled "Perform single cell analysis, integrate with CRISPRa screen"
def mock_integrated_analysis_notebook():
    import scanpy as sc

    # Create a new transform to mimic a new notebook
    # In reality you just run ln.track() in a notebook
    transform = ln.Transform(
        name="Perform single cell analysis, integrate with CRISPRa screen",
        type="notebook",
    )
    transform.save()
    run = ln.Run(transform=transform)
    run.save()

    # access the output files of bfx pipeline and previous analysis
    file_ps = ln.Artifact.filter(description__icontains="perturbseq").one()
    adata = file_ps.load(is_run_input=run)
    file_hits = ln.Artifact.filter(description="hits from schmidt22 crispra GWS").one()
    screen_hits = file_hits.load(is_run_input=run)

    # perform analysis and register output plot files
    sc.tl.score_genes(adata, adata.var_names.intersection(screen_hits.index).tolist())
    filesuffix = "_fig1_score-wgs-hits.png"
    sc.pl.umap(adata, color="score", show=False, save=filesuffix)
    filepath = f"figures/umap{filesuffix}"
    artifact = ln.Artifact(filepath, key=filepath, run=run)
    artifact.save()
    filesuffix = "fig2_score-wgs-hits-per-cluster.png"
    sc.pl.matrixplot(
        adata, groupby="cluster_name", var_names=["score"], show=False, save=filesuffix
    )
    filepath = f"figures/matrixplot_{filesuffix}"
    artifact = ln.Artifact(filepath, key=filepath, run=run)
    artifact.save()

mock_integrated_analysis_notebook()

Show code cell output Hide code cell output

WARNING: saving figure to file figures/umap_fig1_score-wgs-hits.png

WARNING: saving figure to file figures/matrixplot_fig2_score-wgs-hits-per-cluster.png

Review results

Let’s load one of the plots:

# track the current notebook as transform
ln.settings.transform.stem_uid = "1LCd8kco9lZU"
ln.settings.transform.version = "0"
ln.track()

💡 notebook imports: ipython==8.23.0 lamindb==0.70.3 scanpy==1.10.1

💡 saved: Transform(uid='1LCd8kco9lZU6K79', name='Project flow', key='project-flow', version='0', type='notebook', updated_at=2024-04-22 10:27:25 UTC, created_by_id=1)

💡 saved: Run(uid='7HvWOjdBAUthkJUtjJyV', transform_id=7, created_by_id=1)

artifact = ln.Artifact.filter(key__contains="figures/matrixplot").one()
artifact.cache()

Show code cell output Hide code cell output

PosixUPath('/home/runner/work/lamin-usecases/lamin-usecases/docs/mydata/.lamindb/5ARkXuM8SnPEOg93JZph.png')

display(Image(filename=artifact.path))

We see that the image artifact is tracked as an input of the current notebook. The input is highlighted, the notebook follows at the bottom:

artifact.view_lineage()

Alternatively, we can also look at the sequence of transforms:

transform = ln.Transform.search("Project flow", return_queryset=True).first()
transform.parents.df()

	uid	name	key	version	description	type	latest_report_id	source_code_id	reference	reference_type	created_at	updated_at	created_by_id
id
6	Oah4jGcn72kwOhS9	Perform single cell analysis, integrate with C...	None	None	None	notebook	None	None	None	None	2024-04-22 10:27:24.934286+00:00	2024-04-22 10:27:24.934319+00:00	1

transform.view_parents()

Understand runs

We tracked pipeline and notebook runs through run_context, which stores a Transform and a Run record as a global context.

Artifact objects are the inputs and outputs of runs.

What if I don’t want a global context?

Sometimes, we don’t want to create a global run context but manually pass a run when creating an artifact:

run = ln.Run(transform=transform)
ln.Artifact(filepath, run=run)

When does an artifact appear as a run input?

When accessing an artifact via stage(), load() or backed(), two things happen:

1. The current run gets added to artifact.input_of

2. The transform of that artifact gets added as a parent of the current transform

You can then switch off auto-tracking of run inputs if you set ln.settings.track_run_inputs = False: Can I disable tracking run inputs?

You can also track run inputs on a case by case basis via is_run_input=True, e.g., here:

artifact.load(is_run_input=True)

Query by provenance

We can query or search for the notebook that created the artifact:

transform = ln.Transform.search("GWS CRIPSRa analysis", return_queryset=True).first()

And then find all the artifacts created by that notebook:

ln.Artifact.filter(transform=transform).df()

	uid	storage_id	key	suffix	accessor	description	version	size	hash	hash_type	n_objects	n_observations	transform_id	run_id	visibility	key_is_virtual	created_at	updated_at	created_by_id
id
2	T0Bso2o2lFp1D72YmXuP	1	None	.parquet	DataFrame	hits from schmidt22 crispra GWS	None	18368	PihzyuN-FWc-ld6ioxAuPg	md5	None	None	2	2	1	True	2024-04-22 10:27:15.719802+00:00	2024-04-22 10:27:15.719834+00:00	1

Which transform ingested a given artifact?

artifact = ln.Artifact.filter().first()
artifact.transform

Transform(uid='IklsxKrGLRImLZh0', name='Upload GWS CRISPRa result', type='upload', updated_at=2024-04-22 10:27:14 UTC, created_by_id=1)

And which user?

artifact.created_by

User(uid='DzTjkKse', handle='testuser1', name='Test User1', updated_at=2024-04-22 10:27:17 UTC)

Which transforms were created by a given user?

users = ln.User.lookup()

ln.Transform.filter(created_by=users.testuser1).df()

	uid	name	key	version	description	type	latest_report_id	source_code_id	reference	reference_type	created_at	updated_at	created_by_id
id
1	IklsxKrGLRImLZh0	Upload GWS CRISPRa result	None	None	None	upload	None	NaN	None	None	2024-04-22 10:27:14.273844+00:00	2024-04-22 10:27:14.273864+00:00	1
2	AoqJpZR5fTDQWDmg	GWS CRIPSRa analysis	None	None	None	notebook	None	NaN	None	None	2024-04-22 10:27:15.667273+00:00	2024-04-22 10:27:15.667310+00:00	1
3	qCJPkOuZAi9q5zKv	chromium_10x_upload.py	chromium_10x_upload.py	1	None	script	None	3.0	None	None	2024-04-22 10:27:17.721778+00:00	2024-04-22 10:27:18.204983+00:00	1
6	Oah4jGcn72kwOhS9	Perform single cell analysis, integrate with C...	None	None	None	notebook	None	NaN	None	None	2024-04-22 10:27:24.934286+00:00	2024-04-22 10:27:24.934319+00:00	1
7	1LCd8kco9lZU6K79	Project flow	project-flow	0	None	notebook	None	NaN	None	None	2024-04-22 10:27:25.844322+00:00	2024-04-22 10:27:25.844361+00:00	1

Which notebooks were created by a given user?

ln.Transform.filter(created_by=users.testuser1, type="notebook").df()

	uid	name	key	version	description	type	latest_report_id	source_code_id	reference	reference_type	created_at	updated_at	created_by_id
id
2	AoqJpZR5fTDQWDmg	GWS CRIPSRa analysis	None	None	None	notebook	None	None	None	None	2024-04-22 10:27:15.667273+00:00	2024-04-22 10:27:15.667310+00:00	1
6	Oah4jGcn72kwOhS9	Perform single cell analysis, integrate with C...	None	None	None	notebook	None	None	None	None	2024-04-22 10:27:24.934286+00:00	2024-04-22 10:27:24.934319+00:00	1
7	1LCd8kco9lZU6K79	Project flow	project-flow	0	None	notebook	None	None	None	None	2024-04-22 10:27:25.844322+00:00	2024-04-22 10:27:25.844361+00:00	1

We can also view all recent additions to the entire database:

ln.view()

Show code cell output Hide code cell output

Artifact

	uid	storage_id	key	suffix	accessor	description	version	size	hash	hash_type	n_objects	n_observations	transform_id	run_id	visibility	key_is_virtual	created_at	updated_at	created_by_id
id
13	5ARkXuM8SnPEOg93JZph	1	figures/matrixplot_fig2_score-wgs-hits-per-clu...	.png	None	None	None	28814	8zXF_cVwaZnfhmrLbt_0kA	md5	None	None	6	6	1	True	2024-04-22 10:27:25.540623+00:00	2024-04-22 10:27:25.540652+00:00	1
12	yOd8h7IYJiPcsVn21m3I	1	figures/umap_fig1_score-wgs-hits.png	.png	None	None	None	118999	DCFDLUMF-UohaBvkThn0mA	md5	None	None	6	6	1	True	2024-04-22 10:27:25.197568+00:00	2024-04-22 10:27:25.197596+00:00	1
11	0yd8V63gtJmF3dgJ5fBB	1	schmidt22_perturbseq.h5ad	.h5ad	AnnData	perturbseq counts	None	20659936	la7EvqEUMDlug9-rpw-udA	md5	None	None	5	5	1	False	2024-04-22 10:27:23.655543+00:00	2024-04-22 10:27:23.655575+00:00	2
9	XzwrfUB2aHg4gA2ibxAv	1	perturbseq/filtered_feature_bc_matrix/matrix.m...	.mtx.gz	None	None	None	6	1CMzKRaZQPd0YsqeZLW2WA	md5	None	None	4	4	1	False	2024-04-22 10:27:20.945048+00:00	2024-04-22 10:27:20.945068+00:00	2
8	y9QEQXXttDsVMoFcjZX6	1	perturbseq/filtered_feature_bc_matrix/features...	.tsv.gz	None	None	None	6	ju_g0clsk42vS2RhqFRXdQ	md5	None	None	4	4	1	False	2024-04-22 10:27:20.944388+00:00	2024-04-22 10:27:20.944416+00:00	2
7	4zjsgH1BeeA149zyMDHc	1	perturbseq/filtered_feature_bc_matrix/barcodes...	.tsv.gz	None	None	None	6	n3K0imoBddWNvh6-slrRUA	md5	None	None	4	4	1	False	2024-04-22 10:27:20.943479+00:00	2024-04-22 10:27:20.943508+00:00	2
6	XdN42RTBrOalaBBwf1rO	1	fastq/perturbseq_R2_001.fastq.gz	.fastq.gz	None	None	None	6	YV4QFaBVS5k7_jvWw8ckNQ	md5	None	None	3	3	1	False	2024-04-22 10:27:18.214035+00:00	2024-04-22 10:27:18.214054+00:00	1

Run

	uid	transform_id	started_at	finished_at	created_by_id	json	report_id	environment_id	is_consecutive	reference	reference_type	created_at
id
1	EWfQAIrW3wbA1uOGeERV	1	2024-04-22 10:27:14.277397+00:00	NaT	1	None	None	NaN	True	None	None	2024-04-22 10:27:14.277594+00:00
2	iXfKJZMNgbp4ObvxtuIk	2	2024-04-22 10:27:15.669288+00:00	NaT	1	None	None	NaN	None	None	None	2024-04-22 10:27:15.669384+00:00
3	BykoALMl9mp8dzfXUNqm	3	2024-04-22 10:27:17.724368+00:00	2024-04-22 10:27:18.215857+00:00	1	None	None	4.0	None	None	None	2024-04-22 10:27:17.724492+00:00
4	oCsKHG1XuH1jK4o0J1BX	4	2024-04-22 10:27:20.450484+00:00	NaT	2	None	None	NaN	None	None	None	2024-04-22 10:27:20.450607+00:00
5	nCDH4kLNiN2ZhCjuwVuO	5	2024-04-22 10:27:22.669651+00:00	NaT	2	None	None	4.0	None	None	None	2024-04-22 10:27:22.669781+00:00
6	WlJjYCJooGQYcl4Lnkpo	6	2024-04-22 10:27:24.936851+00:00	NaT	1	None	None	NaN	None	None	None	2024-04-22 10:27:24.936952+00:00
7	7HvWOjdBAUthkJUtjJyV	7	2024-04-22 10:27:25.849930+00:00	NaT	1	None	None	NaN	True	None	None	2024-04-22 10:27:25.850031+00:00

Storage

	uid	root	description	type	region	created_at	updated_at	created_by_id
id
1	3vZBLuqy	/home/runner/work/lamin-usecases/lamin-usecase...	None	local	None	2024-04-22 10:27:12.274302+00:00	2024-04-22 10:27:12.274327+00:00	1

Transform

	uid	name	key	version	description	type	latest_report_id	source_code_id	reference	reference_type	created_at	updated_at	created_by_id
id
7	1LCd8kco9lZU6K79	Project flow	project-flow	0	None	notebook	None	NaN	None	None	2024-04-22 10:27:25.844322+00:00	2024-04-22 10:27:25.844361+00:00	1
6	Oah4jGcn72kwOhS9	Perform single cell analysis, integrate with C...	None	None	None	notebook	None	NaN	None	None	2024-04-22 10:27:24.934286+00:00	2024-04-22 10:27:24.934319+00:00	1
5	YqmbO6oMXjRj65cN	postprocess_cellranger.py	postprocess_cellranger.py	2	None	script	None	10.0	None	None	2024-04-22 10:27:22.666862+00:00	2024-04-22 10:27:23.146977+00:00	2
4	w6eKweXO1stJDZiW	Cell Ranger	None	7.2.0	None	pipeline	None	NaN	https://www.10xgenomics.com/support/software/c...	None	2024-04-22 10:27:20.448137+00:00	2024-04-22 10:27:20.448161+00:00	2
3	qCJPkOuZAi9q5zKv	chromium_10x_upload.py	chromium_10x_upload.py	1	None	script	None	3.0	None	None	2024-04-22 10:27:17.721778+00:00	2024-04-22 10:27:18.204983+00:00	1
2	AoqJpZR5fTDQWDmg	GWS CRIPSRa analysis	None	None	None	notebook	None	NaN	None	None	2024-04-22 10:27:15.667273+00:00	2024-04-22 10:27:15.667310+00:00	1
1	IklsxKrGLRImLZh0	Upload GWS CRISPRa result	None	None	None	upload	None	NaN	None	None	2024-04-22 10:27:14.273844+00:00	2024-04-22 10:27:14.273864+00:00	1

User

	uid	handle	name	created_at	updated_at
id
2	bKeW4T6E	testuser2	Test User2	2024-04-22 10:27:15.658644+00:00	2024-04-22 10:27:20.423826+00:00
1	DzTjkKse	testuser1	Test User1	2024-04-22 10:27:12.271119+00:00	2024-04-22 10:27:17.588457+00:00

Show code cell content Hide code cell content

!lamin login testuser1
!lamin delete --force mydata
!rm -r ./mydata

✅ logged in with email testuser1@lamin.ai (uid: DzTjkKse)

💡 deleting instance testuser1/mydata
❗ manually delete your stored data: /home/runner/work/lamin-usecases/lamin-usecases/docs/mydata