Curate & link multi-modal data#

import lamindb as ln
import lnschema_bionty as lb

lb.settings.species = "human"
ln.settings.verbosity = 3

✅ loaded instance: testuser1/test-multimodal (lamindb 0.50.3)

🌱 set species: Species(id='uHJU', name='human', taxon_id=9606, scientific_name='homo_sapiens', updated_at=2023-08-12 05:44:10, bionty_source_id='ZCYZ', created_by_id='DzTjkKse')

ln.track()

💡 notebook imports: lamindb==0.50.3 lnschema_bionty==0.29.3

🌱 saved: Transform(id='yMWSFirS6qv2z8', name='Curate & link multi-modal data', short_name='multimodal', stem_id='yMWSFirS6qv2', version='0', type=notebook, updated_at=2023-08-12 05:44:10, created_by_id='DzTjkKse')

🌱 saved: Run(id='QWLQPw6dtIoUZNwo2gkP', run_at=2023-08-12 05:44:10, transform_id='yMWSFirS6qv2z8', created_by_id='DzTjkKse')

MuData object#

Let’s use a MuData object:

mdata

MuData object with n_obs × n_vars = 200 × 300
  var:	'name'
  4 modalities
    rna:	200 x 173
      obs:	'orig.ident', 'nCount_RNA', 'nFeature_RNA', 'nCount_HTO', 'nFeature_HTO', 'nCount_GDO', 'nCount_ADT', 'nFeature_ADT', 'percent.mito', 'MULTI_ID', 'HTO_classification', 'guide_ID', 'gene_target', 'NT', 'perturbation', 'replicate', 'S.Score', 'G2M.Score', 'Phase'
      var:	'name'
    adt:	200 x 4
      obs:	'orig.ident', 'nCount_RNA', 'nFeature_RNA', 'nCount_HTO', 'nFeature_HTO', 'nCount_GDO', 'nCount_ADT', 'nFeature_ADT', 'percent.mito', 'MULTI_ID', 'HTO_classification', 'guide_ID', 'gene_target', 'NT', 'perturbation', 'replicate', 'S.Score', 'G2M.Score', 'Phase'
      var:	'name'
    hto:	200 x 12
      obs:	'orig.ident', 'nCount_RNA', 'nFeature_RNA', 'nCount_HTO', 'nFeature_HTO', 'nCount_GDO', 'nCount_ADT', 'nFeature_ADT', 'percent.mito', 'MULTI_ID', 'HTO_classification', 'guide_ID', 'gene_target', 'NT', 'perturbation', 'replicate', 'S.Score', 'G2M.Score', 'Phase'
      var:	'name'
    gdo:	200 x 111
      obs:	'orig.ident', 'nCount_RNA', 'nFeature_RNA', 'nCount_HTO', 'nFeature_HTO', 'nCount_GDO', 'nCount_ADT', 'nFeature_ADT', 'percent.mito', 'MULTI_ID', 'HTO_classification', 'guide_ID', 'gene_target', 'NT', 'perturbation', 'replicate', 'S.Score', 'G2M.Score', 'Phase'
      var:	'name'

First we register the file:

file = ln.File(
    "papalexi21_subset.h5mu", description="Sub-sampled MuData from Papalexi21"
)
file.save()

🌱 storing file 'xSaT4QuhgTDxZ2QKBdEP' with key '.lamindb/xSaT4QuhgTDxZ2QKBdEP.h5mu'

Register features#

Now let’s register the 3 feature sets this data contains:

rna
adt
obs (metadata)

modalities#

For the two modalities rna and adt, we use bionty tables as the reference:

mdata["rna"].var_names[:5]

Index(['RP5-827C21.6', 'XX-CR54.1', 'SH2D6', 'RP11-379B18.5', 'RP11-778D9.12'], dtype='object', name='index')

feature_set_rna = ln.FeatureSet.from_values(
    mdata["rna"].var_names, field=lb.Gene.symbol
)

💡 using global setting species = human

✅ validated 93 Gene records from Bionty on symbol: SH2D6, ARHGAP26-AS1, GABRA1, HLA-DQB1-AS1, HLA-DQB1-AS1, HLA-DQB1-AS1, HLA-DQB1-AS1, HLA-DQB1-AS1, HLA-DQB1-AS1, HLA-DQB1-AS1, SPACA1, VNN1, CTAGE15, CTAGE15, PFKFB1, TRPC5, RBPMS-AS1, CA8, CSMD3, ZNF483, ...

🔶 ambiguous validation in Bionty for 11 records: HLA-DQB1-AS1, CTAGE15, CRYAB, CTRB2, LGALS9C, NPHS1, THPO, PCDHB11, XG, TBC1D3G, TUBB1

🔶 did not validate 96 Gene records for symbols: AC002066.1, AC004019.13, AC005150.1, AC006042.7, AC011558.5, AC026471.6, AC073934.6, AC091132.1, AC092295.4, AC092687.5, AE000662.93, AL132989.1, AP000442.4, AP003419.16, C14orf177, C1orf65, CASC1, CTA-373H7.7, CTB-134F13.1, CTB-31O20.9, ...

🔶 ignoring non-validated features: AC002066.1,AC004019.13,AC005150.1,AC006042.7,AC011558.5,AC026471.6,AC073934.6,AC091132.1,AC092295.4,AC092687.5,AE000662.93,AL132989.1,AP000442.4,AP003419.16,C14orf177,C1orf65,CASC1,CTA-373H7.7,CTB-134F13.1,CTB-31O20.9,CTC-467M3.1,CTC-498J12.1,CTD-2562J17.2,CTD-3012A18.1,CTD-3065B20.2,CTD-3193O13.8,FAM65C,HIST1H4K,IBA57-AS1,KIAA1239,LARGE,NBPF16,RP1-1J6.2,RP11-110I1.14,RP11-113K21.4,RP11-120C12.3,RP11-12D24.10,RP11-12J10.3,RP11-134K13.4,RP11-136I14.5,RP11-138C9.1,RP11-146I2.1,RP11-152H18.3,RP11-17J14.2,RP11-186N15.3,RP11-187A9.3,RP11-214C8.2,RP11-219B4.7,RP11-231C14.4,RP11-235C23.5,RP11-247A12.8,RP11-265N6.2,RP11-268G12.1,RP11-2H8.2,RP11-304L19.11,RP11-307N16.6,RP11-324E6.9,RP11-325L7.1,RP11-32B5.7,RP11-335O4.3,RP11-346D14.1,RP11-365N19.2,RP11-379B18.5,RP11-3D4.2,RP11-403P17.5,RP11-408E5.4,RP11-415J8.5,RP11-434D9.1,RP11-465N4.4,RP11-473O4.4,RP11-496I9.1,RP11-524H19.2,RP11-532F6.4,RP11-536K7.5,RP11-546K22.3,RP11-624M8.1,RP11-703G6.1,RP11-717H13.1,RP11-745O10.2,RP11-75C10.9,RP11-760N9.1,RP11-778D9.12,RP11-80H5.7,RP11-835E18.5,RP11-867G23.3,RP11-973N13.4,RP11-982M15.2,RP11-9M16.2,RP13-582O9.7,RP3-327A19.5,RP3-337O18.9,RP5-827C21.6,RP5-855F16.1,TMEM75,U52111.14,XX-CR54.1

mdata["adt"].var_names

Index(['CD86', 'PDL1', 'PDL2', 'CD366'], dtype='object', name='index')

feature_set_adt = ln.FeatureSet.from_values(
    mdata["adt"].var_names, field=lb.CellMarker.name
)

💡 using global setting species = human

✅ validated 4 CellMarker records from Bionty on name: CD86, PDL1, PDL2, CD366

Link them to file:

file.features.add_feature_set(feature_set_rna, slot="rna")
file.features.add_feature_set(feature_set_adt, slot="adt")

metadata#

The 3rd feature set is the obs:

obs = mdata["rna"].obs

We’re only interested in a single metadata column:

ln.Feature(name="gene_target", type="category").save()

feature_set_obs = ln.FeatureSet.from_df(obs, "metadata")

✅ validated 1 Feature record on name: gene_target

🔶 did not validate 18 Feature records for names: G2M.Score, HTO_classification, MULTI_ID, NT, Phase, S.Score, guide_ID, nCount_ADT, nCount_GDO, nCount_HTO, nCount_RNA, nFeature_ADT, nFeature_HTO, nFeature_RNA, orig.ident, percent.mito, perturbation, replicate

🔶 ignoring non-validated features: G2M.Score,HTO_classification,MULTI_ID,NT,Phase,S.Score,guide_ID,nCount_ADT,nCount_GDO,nCount_HTO,nCount_RNA,nFeature_ADT,nFeature_HTO,nFeature_RNA,orig.ident,percent.mito,perturbation,replicate

file.features.add_feature_set(feature_set_obs, slot="obs")

gene_targets = lb.Gene.from_values(obs["gene_target"], "symbol")
ln.save(gene_targets)
file.add_labels(gene_targets)

💡 using global setting species = human

✅ validated 35 Gene records from Bionty on symbol: IFNGR1, IFNGR1, CAV1, IRF7, IRF7, IRF7, ATF2, NFKBIA, NFKBIA, STAT1, STAT1, SPI1, JAK2, JAK2, STAT2, STAT2, IFNGR2, IFNGR2, IFNGR2, CD86, ...

🔶 ambiguous validation in Bionty for 10 records: IFNGR1, IRF7, NFKBIA, STAT1, JAK2, STAT2, IFNGR2, SMAD4, STAT3, TNFRSF14

🔶 did not validate 2 Gene records for symbols: MARCH8, NT

🌱 linked labels 'IFNGR1', 'IFNGR1', 'CAV1', 'IRF7', 'IRF7', 'IRF7', 'ATF2', 'NFKBIA', 'NFKBIA', 'STAT1', 'STAT1', 'SPI1', 'JAK2', 'JAK2', 'STAT2', 'STAT2', 'IFNGR2', 'IFNGR2', 'IFNGR2', 'CD86', 'STAT5A', 'SMAD4', 'SMAD4', 'ETV7', 'IRF1', 'UBE2L6', 'PDCD1LG2', 'BRD4', 'POU2F2', 'STAT3', 'STAT3', 'TNFRSF14', 'TNFRSF14', 'CUL3', 'CMTM6', 'MARCH8', 'NT' to feature 'gene_target', linked feature 'gene_target' to registry 'bionty.Gene'

labels = []
for col in ["orig.ident", "perturbation", "replicate", "Phase", "guide_ID"]:
    labels += ln.Label.from_values(obs[col])

🔶 did not validate 8 Label records for names: Lane7, Lane4, Lane2, Lane5, Lane3, Lane8, Lane1, Lane6

🔶 did not validate 2 Label records for names: Perturbed, NT

🔶 did not validate 3 Label records for names: rep3, rep1, rep2

🔶 did not validate 3 Label records for names: G1, S, G2M

🔶 did not validate 78 Label records for names: MARCH8g2, IFNGR1g3, MARCH8g4, CAV1g4, IRF7g1, ATF2g1, NFKBIAg2, STAT1g2, SPI1g1, JAK2g3, NTg7, IFNGR1g4, NTg1, STAT2g2, IFNGR2g2, CD86g2, IFNGR2g1, STAT5Ag2, IFNGR1g2, IFNGR1g1, ...

Because none of these labels seem like something we’d want to track in the registry or validate, we don’t link them to the file.

file.features

'rna': FeatureSet(id='210mjWwxiIbXls7NIQPp', n=93, type='float', registry='bionty.Gene', hash='6Sd_y8RL6Uy6JQCuHM6Y', updated_at=2023-08-12 05:44:13, created_by_id='DzTjkKse')
'adt': FeatureSet(id='yl73Vq8Kpq6GpmxiXlXS', n=4, type='float', registry='bionty.CellMarker', hash='b-CtyjgPRO0WN27lTOqC', updated_at=2023-08-12 05:44:13, created_by_id='DzTjkKse')
'obs': FeatureSet(id='ZEHCV0sPHa9YnbLNHx4K', name='metadata', n=1, registry='core.Feature', hash='ODo4T8yNUUqlmCysIaMH', updated_at=2023-08-12 05:44:13, created_by_id='DzTjkKse')

file.describe()

💡 File(id=xSaT4QuhgTDxZ2QKBdEP, key=None, suffix=.h5mu, accessor=MuData, description=Sub-sampled MuData from Papalexi21, version=None, size=606320, hash=RaivS3NesDOP-6kNIuaC3g, hash_type=md5, created_at=2023-08-12 05:44:10.955330+00:00, updated_at=2023-08-12 05:44:10.955354+00:00)

Provenance:
    🗃️ storage: Storage(id='swspYvOl', root='/home/runner/work/lamin-usecases/lamin-usecases/docs/test-multimodal', type='local', updated_at=2023-08-12 05:44:08, created_by_id='DzTjkKse')
    📎 initial_version: None
    📔 transform: Transform(id='yMWSFirS6qv2z8', name='Curate & link multi-modal data', short_name='multimodal', stem_id='yMWSFirS6qv2', version='0', type='notebook', updated_at=2023-08-12 05:44:10, created_by_id='DzTjkKse')
    🚗 run: Run(id='QWLQPw6dtIoUZNwo2gkP', run_at=2023-08-12 05:44:10, transform_id='yMWSFirS6qv2z8', created_by_id='DzTjkKse')
    👤 created_by: User(id='DzTjkKse', handle='testuser1', email='testuser1@lamin.ai', name='Test User1', updated_at=2023-08-12 05:44:08)
Features:
  🗺️ rna:
    🔗 index (93, bionty.Gene.id): ['I02NZSZmvIzj', 'dQgCTEHCW6eg', 'q017r7kC52Mu', 'nsBNCWQPmYZq', 'wE77gLNzdGJw'...]
  🗺️ adt:
    🔗 index (4, bionty.CellMarker.id): ['82nG0xqSuEQD', 'BK30rjK34sZd', 'L0m6f7FPiDeg', 'kbrA7wdDuqDK'...]
  🗺️ obs (metadata):
    🔗 gene_target (37, bionty.Gene): ['STAT1', 'IRF7', 'SPI1', 'UBE2L6', 'IFNGR2']

file.view_lineage()

https://d33wubrfki0l68.cloudfront.net/6d3a6a62c982ab07052a4db952319cac5d50ab01/725d3/_images/a0b4f1a6cdf7a5458f90d91ebe3de2c046b3ceb19645d1560f1a1eba67b2c031.svg