Dandelion class

Much of the functions and utility of the dandelion package revolves around the Dandelion class object. The class will act as an intermediary object for storage and flexible interaction with other tools. This section will run through a quick primer to the Dandelion class.

Import modules

[1]:

import os

os.chdir(os.path.expanduser("~/Downloads/dandelion_tutorial/"))
import dandelion as ddl

ddl.logging.print_versions()

dandelion==0.3.4.dev30 pandas==2.0.1 numpy==1.24.3 matplotlib==3.7.1 networkx==3.1 scipy==1.11.2

[2]:

vdj = ddl.read_h5ddl("dandelion_results.h5ddl")
vdj

[2]:

Dandelion class object with n_obs = 2071 and n_contigs = 4882
    data: 'sequence_id', 'sequence', 'rev_comp', 'productive', 'v_call', 'd_call', 'j_call', 'sequence_alignment', 'germline_alignment', 'junction', 'junction_aa', 'v_cigar', 'd_cigar', 'j_cigar', 'stop_codon', 'vj_in_frame', 'locus', 'c_call', 'junction_length', 'np1_length', 'np2_length', 'v_sequence_start', 'v_sequence_end', 'v_germline_start', 'v_germline_end', 'd_sequence_start', 'd_sequence_end', 'd_germline_start', 'd_germline_end', 'j_sequence_start', 'j_sequence_end', 'j_germline_start', 'j_germline_end', 'v_score', 'v_identity', 'v_support', 'd_score', 'd_identity', 'd_support', 'j_score', 'j_identity', 'j_support', 'fwr1', 'fwr2', 'fwr3', 'fwr4', 'cdr1', 'cdr2', 'cdr3', 'cell_id', 'consensus_count', 'duplicate_count', 'v_call_10x', 'd_call_10x', 'j_call_10x', 'junction_10x', 'junction_10x_aa', 'j_support_igblastn', 'j_score_igblastn', 'j_call_igblastn', 'j_call_blastn', 'j_identity_blastn', 'j_alignment_length_blastn', 'j_number_of_mismatches_blastn', 'j_number_of_gap_openings_blastn', 'j_sequence_start_blastn', 'j_sequence_end_blastn', 'j_germline_start_blastn', 'j_germline_end_blastn', 'j_support_blastn', 'j_score_blastn', 'j_sequence_alignment_blastn', 'j_germline_alignment_blastn', 'j_source', 'd_support_igblastn', 'd_score_igblastn', 'd_call_igblastn', 'd_call_blastn', 'd_identity_blastn', 'd_alignment_length_blastn', 'd_number_of_mismatches_blastn', 'd_number_of_gap_openings_blastn', 'd_sequence_start_blastn', 'd_sequence_end_blastn', 'd_germline_start_blastn', 'd_germline_end_blastn', 'd_support_blastn', 'd_score_blastn', 'd_sequence_alignment_blastn', 'd_germline_alignment_blastn', 'd_source', 'v_call_genotyped', 'germline_alignment_d_mask', 'sample_id', 'c_sequence_alignment', 'c_germline_alignment', 'c_sequence_start', 'c_sequence_end', 'c_score', 'c_identity', 'c_call_10x', 'junction_aa_length', 'fwr1_aa', 'fwr2_aa', 'fwr3_aa', 'fwr4_aa', 'cdr1_aa', 'cdr2_aa', 'cdr3_aa', 'sequence_alignment_aa', 'v_sequence_alignment_aa', 'd_sequence_alignment_aa', 'j_sequence_alignment_aa', 'complete_vdj', 'j_call_multimappers', 'j_call_multiplicity', 'j_call_sequence_start_multimappers', 'j_call_sequence_end_multimappers', 'j_call_support_multimappers', 'mu_count', 'ambiguous', 'rearrangement_status', 'clone_id', 'changeo_clone_id'
    metadata: 'clone_id', 'clone_id_by_size', 'sample_id', 'locus_VDJ', 'locus_VJ', 'productive_VDJ', 'productive_VJ', 'v_call_genotyped_VDJ', 'd_call_VDJ', 'j_call_VDJ', 'v_call_genotyped_VJ', 'j_call_VJ', 'c_call_VDJ', 'c_call_VJ', 'junction_VDJ', 'junction_VJ', 'junction_aa_VDJ', 'junction_aa_VJ', 'v_call_genotyped_B_VDJ', 'd_call_B_VDJ', 'j_call_B_VDJ', 'v_call_genotyped_B_VJ', 'j_call_B_VJ', 'c_call_B_VDJ', 'c_call_B_VJ', 'productive_B_VDJ', 'productive_B_VJ', 'duplicate_count_B_VDJ', 'duplicate_count_B_VJ', 'v_call_VDJ_main', 'v_call_VJ_main', 'd_call_VDJ_main', 'j_call_VDJ_main', 'j_call_VJ_main', 'c_call_VDJ_main', 'c_call_VJ_main', 'v_call_B_VDJ_main', 'd_call_B_VDJ_main', 'j_call_B_VDJ_main', 'v_call_B_VJ_main', 'j_call_B_VJ_main', 'isotype', 'isotype_status', 'locus_status', 'chain_status', 'rearrangement_status_VDJ', 'rearrangement_status_VJ', 'changeo_clone_id'
    layout: layout for 2071 vertices, layout for 70 vertices
    graph: networkx graph of 2071 vertices, networkx graph of 70 vertices

Essentially, the .data slot holds the AIRR contig table while the .metadata holds a collapsed version that is compatible with combining with AnnData’s .obs slot. You can retrieve these slots like a typical class object; for example, if I want the metadata:

[3]:

vdj.metadata

[3]:

	clone_id	clone_id_by_size	sample_id	locus_VDJ	locus_VJ	productive_VDJ	productive_VJ	v_call_genotyped_VDJ	d_call_VDJ	j_call_VDJ	...	j_call_B_VDJ_main	v_call_B_VJ_main	j_call_B_VJ_main	isotype	isotype_status	locus_status	chain_status	rearrangement_status_VDJ	rearrangement_status_VJ	changeo_clone_id
sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC	B_VDJ_119_3_2_VJ_80_2_3	1952	sc5p_v2_hs_PBMC_10k	IGH	IGK	T	T	IGHV1-69,IGHV1-69D	IGHD3-22	IGHJ3	...	IGHJ3	IGKV1-8	IGKJ1	IgM	IgM	IGH + IGK	Single pair	standard	standard	11_0
sc5p_v2_hs_PBMC_10k_AAACCTGTCGAGAACG	B_VDJ_42_1_2_VJ_54_1_1	1567	sc5p_v2_hs_PBMC_10k	IGH	IGL	T	T	IGHV1-2	None	IGHJ3	...	IGHJ3	IGLV5-45	IGLJ3	IgM	IgM	IGH + IGL	Single pair	standard	standard	150_1
sc5p_v2_hs_PBMC_10k_AAACCTGTCTTGAGAC	B_VDJ_38_4_4_VJ_191_1_1	1568	sc5p_v2_hs_PBMC_10k	IGH	IGK	T	T	IGHV5-51	None	IGHJ3	...	IGHJ3	IGKV1D-8	IGKJ2	IgM	IgM	IGH + IGK	Single pair	standard	standard	322_2
sc5p_v2_hs_PBMC_10k_AAACGGGAGCGACGTA	B_VDJ_55_2_1_VJ_184_2_7	1569	sc5p_v2_hs_PBMC_10k	IGH	IGL	T	T	IGHV4-4	IGHD6-13	IGHJ3	...	IGHJ3	IGLV3-19	IGLJ3,IGLJ2	IgM	IgM	IGH + IGL	Single pair	standard	standard	290_3
sc5p_v2_hs_PBMC_10k_AAACGGGCACTGTTAG	B_VDJ_44_2_3_VJ_164_3_4	1570	sc5p_v2_hs_PBMC_10k	IGH	IGL	T	T	IGHV4-39	IGHD3-22	IGHJ3	...	IGHJ3	IGLV3-21	IGLJ3,IGLJ2	IgM	IgM	IGH + IGL	Single pair	standard	standard	518_4
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
vdj_v1_hs_pbmc3_TTTCCTCAGCAATATG	B_VDJ_41_2_1_VJ_26_2_8	796	vdj_v1_hs_pbmc3	IGH	IGK	T	T	IGHV2-5	IGHD5/OR15-5a,IGHD5/OR15-5b	IGHJ5,IGHJ4	...	IGHJ5,IGHJ4	IGKV4-1	IGKJ4	IgM	IgM	IGH + IGK	Single pair	standard	standard	1762_1974
vdj_v1_hs_pbmc3_TTTCCTCAGCGCTTAT	B_VDJ_2_6_3_VJ_87_1_3	797	vdj_v1_hs_pbmc3	IGH	IGK	T	T	IGHV3-30	IGHD4-17	IGHJ6	...	IGHJ6	IGKV2-30	IGKJ2	IgM	IgM	IGH + IGK	Single pair	standard	standard	833_1975
vdj_v1_hs_pbmc3_TTTCCTCAGGGAAACA	B_VDJ_1_1_1_VJ_139_4_16	798	vdj_v1_hs_pbmc3	IGH	IGK	T	T	IGHV4-59	IGHD6-13	IGHJ2	...	IGHJ2	IGKV1D-39,IGKV1-39	IGKJ1	IgM	IgM	IGH + IGK	Single pair	standard	standard	1203_1976
vdj_v1_hs_pbmc3_TTTGCGCCATACCATG	B_VDJ_47_4_1_VJ_103_3_4	799	vdj_v1_hs_pbmc3	IGH	IGL	T	T	IGHV1-69,IGHV1-69D	IGHD2-15	IGHJ6	...	IGHJ6	IGLV1-47	IGLJ3	IgM	IgM	IGH + IGL	Single pair	standard	standard	1803_1977
vdj_v1_hs_pbmc3_TTTGGTTGTAGGCATG	B_VDJ_184_5_1_VJ_121_3_3	2336	vdj_v1_hs_pbmc3	IGH	IGL	T	T	IGHV3-23,IGHV3-23D	None	IGHJ4	...	IGHJ4	IGLV2-11	IGLJ3,IGLJ2	IgM	IgM	IGH + IGL	Single pair	standard	standard	1937_1978

2071 rows × 48 columns

slicing

You can slice the Dandelion object via the .data or .metadata via their indices, with the behavior similar to how it is in pandas DataFrame and AnnData.

slicing .data

[4]:

vdj[vdj.data["clone_id"] == "B_VDJ_41_2_1_VJ_26_2_8"]

[4]:

Dandelion class object with n_obs = 1 and n_contigs = 2
    data: 'sequence_id', 'sequence', 'rev_comp', 'productive', 'v_call', 'd_call', 'j_call', 'sequence_alignment', 'germline_alignment', 'junction', 'junction_aa', 'v_cigar', 'd_cigar', 'j_cigar', 'stop_codon', 'vj_in_frame', 'locus', 'c_call', 'junction_length', 'np1_length', 'np2_length', 'v_sequence_start', 'v_sequence_end', 'v_germline_start', 'v_germline_end', 'd_sequence_start', 'd_sequence_end', 'd_germline_start', 'd_germline_end', 'j_sequence_start', 'j_sequence_end', 'j_germline_start', 'j_germline_end', 'v_score', 'v_identity', 'v_support', 'd_score', 'd_identity', 'd_support', 'j_score', 'j_identity', 'j_support', 'fwr1', 'fwr2', 'fwr3', 'fwr4', 'cdr1', 'cdr2', 'cdr3', 'cell_id', 'consensus_count', 'duplicate_count', 'v_call_10x', 'd_call_10x', 'j_call_10x', 'junction_10x', 'junction_10x_aa', 'j_support_igblastn', 'j_score_igblastn', 'j_call_igblastn', 'j_call_blastn', 'j_identity_blastn', 'j_alignment_length_blastn', 'j_number_of_mismatches_blastn', 'j_number_of_gap_openings_blastn', 'j_sequence_start_blastn', 'j_sequence_end_blastn', 'j_germline_start_blastn', 'j_germline_end_blastn', 'j_support_blastn', 'j_score_blastn', 'j_sequence_alignment_blastn', 'j_germline_alignment_blastn', 'j_source', 'd_support_igblastn', 'd_score_igblastn', 'd_call_igblastn', 'd_call_blastn', 'd_identity_blastn', 'd_alignment_length_blastn', 'd_number_of_mismatches_blastn', 'd_number_of_gap_openings_blastn', 'd_sequence_start_blastn', 'd_sequence_end_blastn', 'd_germline_start_blastn', 'd_germline_end_blastn', 'd_support_blastn', 'd_score_blastn', 'd_sequence_alignment_blastn', 'd_germline_alignment_blastn', 'd_source', 'v_call_genotyped', 'germline_alignment_d_mask', 'sample_id', 'c_sequence_alignment', 'c_germline_alignment', 'c_sequence_start', 'c_sequence_end', 'c_score', 'c_identity', 'c_call_10x', 'junction_aa_length', 'fwr1_aa', 'fwr2_aa', 'fwr3_aa', 'fwr4_aa', 'cdr1_aa', 'cdr2_aa', 'cdr3_aa', 'sequence_alignment_aa', 'v_sequence_alignment_aa', 'd_sequence_alignment_aa', 'j_sequence_alignment_aa', 'complete_vdj', 'j_call_multimappers', 'j_call_multiplicity', 'j_call_sequence_start_multimappers', 'j_call_sequence_end_multimappers', 'j_call_support_multimappers', 'mu_count', 'ambiguous', 'rearrangement_status', 'clone_id', 'changeo_clone_id'
    metadata: 'clone_id', 'clone_id_by_size', 'sample_id', 'locus_VDJ', 'locus_VJ', 'productive_VDJ', 'productive_VJ', 'v_call_genotyped_VDJ', 'd_call_VDJ', 'j_call_VDJ', 'v_call_genotyped_VJ', 'j_call_VJ', 'c_call_VDJ', 'c_call_VJ', 'junction_VDJ', 'junction_VJ', 'junction_aa_VDJ', 'junction_aa_VJ', 'v_call_genotyped_B_VDJ', 'd_call_B_VDJ', 'j_call_B_VDJ', 'v_call_genotyped_B_VJ', 'j_call_B_VJ', 'c_call_B_VDJ', 'c_call_B_VJ', 'productive_B_VDJ', 'productive_B_VJ', 'duplicate_count_B_VDJ', 'duplicate_count_B_VJ', 'v_call_VDJ_main', 'v_call_VJ_main', 'd_call_VDJ_main', 'j_call_VDJ_main', 'j_call_VJ_main', 'c_call_VDJ_main', 'c_call_VJ_main', 'v_call_B_VDJ_main', 'd_call_B_VDJ_main', 'j_call_B_VDJ_main', 'v_call_B_VJ_main', 'j_call_B_VJ_main', 'isotype', 'isotype_status', 'locus_status', 'chain_status', 'rearrangement_status_VDJ', 'rearrangement_status_VJ', 'changeo_clone_id'
    layout: layout for 1 vertices, layout for 0 vertices
    graph: networkx graph of 1 vertices, networkx graph of 0 vertices

[5]:

vdj[
    vdj.data_names.isin(
        [
            "sc5p_v2_hs_PBMC_10k_AAACCTGTCATATCGG_contig_1",
            "sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC_contig_2",
            "sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC_contig_1",
            "sc5p_v2_hs_PBMC_10k_AAACCTGTCGAGAACG_contig_1",
            "sc5p_v2_hs_PBMC_10k_AAACCTGTCGAGAACG_contig_2",
        ]
    )
]

[5]:

Dandelion class object with n_obs = 2 and n_contigs = 4
    data: 'sequence_id', 'sequence', 'rev_comp', 'productive', 'v_call', 'd_call', 'j_call', 'sequence_alignment', 'germline_alignment', 'junction', 'junction_aa', 'v_cigar', 'd_cigar', 'j_cigar', 'stop_codon', 'vj_in_frame', 'locus', 'c_call', 'junction_length', 'np1_length', 'np2_length', 'v_sequence_start', 'v_sequence_end', 'v_germline_start', 'v_germline_end', 'd_sequence_start', 'd_sequence_end', 'd_germline_start', 'd_germline_end', 'j_sequence_start', 'j_sequence_end', 'j_germline_start', 'j_germline_end', 'v_score', 'v_identity', 'v_support', 'd_score', 'd_identity', 'd_support', 'j_score', 'j_identity', 'j_support', 'fwr1', 'fwr2', 'fwr3', 'fwr4', 'cdr1', 'cdr2', 'cdr3', 'cell_id', 'consensus_count', 'duplicate_count', 'v_call_10x', 'd_call_10x', 'j_call_10x', 'junction_10x', 'junction_10x_aa', 'j_support_igblastn', 'j_score_igblastn', 'j_call_igblastn', 'j_call_blastn', 'j_identity_blastn', 'j_alignment_length_blastn', 'j_number_of_mismatches_blastn', 'j_number_of_gap_openings_blastn', 'j_sequence_start_blastn', 'j_sequence_end_blastn', 'j_germline_start_blastn', 'j_germline_end_blastn', 'j_support_blastn', 'j_score_blastn', 'j_sequence_alignment_blastn', 'j_germline_alignment_blastn', 'j_source', 'd_support_igblastn', 'd_score_igblastn', 'd_call_igblastn', 'd_call_blastn', 'd_identity_blastn', 'd_alignment_length_blastn', 'd_number_of_mismatches_blastn', 'd_number_of_gap_openings_blastn', 'd_sequence_start_blastn', 'd_sequence_end_blastn', 'd_germline_start_blastn', 'd_germline_end_blastn', 'd_support_blastn', 'd_score_blastn', 'd_sequence_alignment_blastn', 'd_germline_alignment_blastn', 'd_source', 'v_call_genotyped', 'germline_alignment_d_mask', 'sample_id', 'c_sequence_alignment', 'c_germline_alignment', 'c_sequence_start', 'c_sequence_end', 'c_score', 'c_identity', 'c_call_10x', 'junction_aa_length', 'fwr1_aa', 'fwr2_aa', 'fwr3_aa', 'fwr4_aa', 'cdr1_aa', 'cdr2_aa', 'cdr3_aa', 'sequence_alignment_aa', 'v_sequence_alignment_aa', 'd_sequence_alignment_aa', 'j_sequence_alignment_aa', 'complete_vdj', 'j_call_multimappers', 'j_call_multiplicity', 'j_call_sequence_start_multimappers', 'j_call_sequence_end_multimappers', 'j_call_support_multimappers', 'mu_count', 'ambiguous', 'rearrangement_status', 'clone_id', 'changeo_clone_id'
    metadata: 'clone_id', 'clone_id_by_size', 'sample_id', 'locus_VDJ', 'locus_VJ', 'productive_VDJ', 'productive_VJ', 'v_call_genotyped_VDJ', 'd_call_VDJ', 'j_call_VDJ', 'v_call_genotyped_VJ', 'j_call_VJ', 'c_call_VDJ', 'c_call_VJ', 'junction_VDJ', 'junction_VJ', 'junction_aa_VDJ', 'junction_aa_VJ', 'v_call_genotyped_B_VDJ', 'd_call_B_VDJ', 'j_call_B_VDJ', 'v_call_genotyped_B_VJ', 'j_call_B_VJ', 'c_call_B_VDJ', 'c_call_B_VJ', 'productive_B_VDJ', 'productive_B_VJ', 'duplicate_count_B_VDJ', 'duplicate_count_B_VJ', 'v_call_VDJ_main', 'v_call_VJ_main', 'd_call_VDJ_main', 'j_call_VDJ_main', 'j_call_VJ_main', 'c_call_VDJ_main', 'c_call_VJ_main', 'v_call_B_VDJ_main', 'd_call_B_VDJ_main', 'j_call_B_VDJ_main', 'v_call_B_VJ_main', 'j_call_B_VJ_main', 'isotype', 'isotype_status', 'locus_status', 'chain_status', 'rearrangement_status_VDJ', 'rearrangement_status_VJ', 'changeo_clone_id'
    layout: layout for 2 vertices, layout for 0 vertices
    graph: networkx graph of 2 vertices, networkx graph of 0 vertices

slicing .metadata

[6]:

vdj[vdj.metadata["productive_VDJ"].isin(["T", "T|T"])]

[6]:

Dandelion class object with n_obs = 2070 and n_contigs = 4875
    data: 'sequence_id', 'sequence', 'rev_comp', 'productive', 'v_call', 'd_call', 'j_call', 'sequence_alignment', 'germline_alignment', 'junction', 'junction_aa', 'v_cigar', 'd_cigar', 'j_cigar', 'stop_codon', 'vj_in_frame', 'locus', 'c_call', 'junction_length', 'np1_length', 'np2_length', 'v_sequence_start', 'v_sequence_end', 'v_germline_start', 'v_germline_end', 'd_sequence_start', 'd_sequence_end', 'd_germline_start', 'd_germline_end', 'j_sequence_start', 'j_sequence_end', 'j_germline_start', 'j_germline_end', 'v_score', 'v_identity', 'v_support', 'd_score', 'd_identity', 'd_support', 'j_score', 'j_identity', 'j_support', 'fwr1', 'fwr2', 'fwr3', 'fwr4', 'cdr1', 'cdr2', 'cdr3', 'cell_id', 'consensus_count', 'duplicate_count', 'v_call_10x', 'd_call_10x', 'j_call_10x', 'junction_10x', 'junction_10x_aa', 'j_support_igblastn', 'j_score_igblastn', 'j_call_igblastn', 'j_call_blastn', 'j_identity_blastn', 'j_alignment_length_blastn', 'j_number_of_mismatches_blastn', 'j_number_of_gap_openings_blastn', 'j_sequence_start_blastn', 'j_sequence_end_blastn', 'j_germline_start_blastn', 'j_germline_end_blastn', 'j_support_blastn', 'j_score_blastn', 'j_sequence_alignment_blastn', 'j_germline_alignment_blastn', 'j_source', 'd_support_igblastn', 'd_score_igblastn', 'd_call_igblastn', 'd_call_blastn', 'd_identity_blastn', 'd_alignment_length_blastn', 'd_number_of_mismatches_blastn', 'd_number_of_gap_openings_blastn', 'd_sequence_start_blastn', 'd_sequence_end_blastn', 'd_germline_start_blastn', 'd_germline_end_blastn', 'd_support_blastn', 'd_score_blastn', 'd_sequence_alignment_blastn', 'd_germline_alignment_blastn', 'd_source', 'v_call_genotyped', 'germline_alignment_d_mask', 'sample_id', 'c_sequence_alignment', 'c_germline_alignment', 'c_sequence_start', 'c_sequence_end', 'c_score', 'c_identity', 'c_call_10x', 'junction_aa_length', 'fwr1_aa', 'fwr2_aa', 'fwr3_aa', 'fwr4_aa', 'cdr1_aa', 'cdr2_aa', 'cdr3_aa', 'sequence_alignment_aa', 'v_sequence_alignment_aa', 'd_sequence_alignment_aa', 'j_sequence_alignment_aa', 'complete_vdj', 'j_call_multimappers', 'j_call_multiplicity', 'j_call_sequence_start_multimappers', 'j_call_sequence_end_multimappers', 'j_call_support_multimappers', 'mu_count', 'ambiguous', 'rearrangement_status', 'clone_id', 'changeo_clone_id'
    metadata: 'clone_id', 'clone_id_by_size', 'sample_id', 'locus_VDJ', 'locus_VJ', 'productive_VDJ', 'productive_VJ', 'v_call_genotyped_VDJ', 'd_call_VDJ', 'j_call_VDJ', 'v_call_genotyped_VJ', 'j_call_VJ', 'c_call_VDJ', 'c_call_VJ', 'junction_VDJ', 'junction_VJ', 'junction_aa_VDJ', 'junction_aa_VJ', 'v_call_genotyped_B_VDJ', 'd_call_B_VDJ', 'j_call_B_VDJ', 'v_call_genotyped_B_VJ', 'j_call_B_VJ', 'c_call_B_VDJ', 'c_call_B_VJ', 'productive_B_VDJ', 'productive_B_VJ', 'duplicate_count_B_VDJ', 'duplicate_count_B_VJ', 'v_call_VDJ_main', 'v_call_VJ_main', 'd_call_VDJ_main', 'j_call_VDJ_main', 'j_call_VJ_main', 'c_call_VDJ_main', 'c_call_VJ_main', 'v_call_B_VDJ_main', 'd_call_B_VDJ_main', 'j_call_B_VDJ_main', 'v_call_B_VJ_main', 'j_call_B_VJ_main', 'isotype', 'isotype_status', 'locus_status', 'chain_status', 'rearrangement_status_VDJ', 'rearrangement_status_VJ', 'changeo_clone_id'
    layout: layout for 2070 vertices, layout for 70 vertices
    graph: networkx graph of 2070 vertices, networkx graph of 70 vertices

[7]:

vdj[vdj.metadata_names == "vdj_v1_hs_pbmc3_TTTCCTCAGCGCTTAT"]

[7]:

Dandelion class object with n_obs = 1 and n_contigs = 2
    data: 'sequence_id', 'sequence', 'rev_comp', 'productive', 'v_call', 'd_call', 'j_call', 'sequence_alignment', 'germline_alignment', 'junction', 'junction_aa', 'v_cigar', 'd_cigar', 'j_cigar', 'stop_codon', 'vj_in_frame', 'locus', 'c_call', 'junction_length', 'np1_length', 'np2_length', 'v_sequence_start', 'v_sequence_end', 'v_germline_start', 'v_germline_end', 'd_sequence_start', 'd_sequence_end', 'd_germline_start', 'd_germline_end', 'j_sequence_start', 'j_sequence_end', 'j_germline_start', 'j_germline_end', 'v_score', 'v_identity', 'v_support', 'd_score', 'd_identity', 'd_support', 'j_score', 'j_identity', 'j_support', 'fwr1', 'fwr2', 'fwr3', 'fwr4', 'cdr1', 'cdr2', 'cdr3', 'cell_id', 'consensus_count', 'duplicate_count', 'v_call_10x', 'd_call_10x', 'j_call_10x', 'junction_10x', 'junction_10x_aa', 'j_support_igblastn', 'j_score_igblastn', 'j_call_igblastn', 'j_call_blastn', 'j_identity_blastn', 'j_alignment_length_blastn', 'j_number_of_mismatches_blastn', 'j_number_of_gap_openings_blastn', 'j_sequence_start_blastn', 'j_sequence_end_blastn', 'j_germline_start_blastn', 'j_germline_end_blastn', 'j_support_blastn', 'j_score_blastn', 'j_sequence_alignment_blastn', 'j_germline_alignment_blastn', 'j_source', 'd_support_igblastn', 'd_score_igblastn', 'd_call_igblastn', 'd_call_blastn', 'd_identity_blastn', 'd_alignment_length_blastn', 'd_number_of_mismatches_blastn', 'd_number_of_gap_openings_blastn', 'd_sequence_start_blastn', 'd_sequence_end_blastn', 'd_germline_start_blastn', 'd_germline_end_blastn', 'd_support_blastn', 'd_score_blastn', 'd_sequence_alignment_blastn', 'd_germline_alignment_blastn', 'd_source', 'v_call_genotyped', 'germline_alignment_d_mask', 'sample_id', 'c_sequence_alignment', 'c_germline_alignment', 'c_sequence_start', 'c_sequence_end', 'c_score', 'c_identity', 'c_call_10x', 'junction_aa_length', 'fwr1_aa', 'fwr2_aa', 'fwr3_aa', 'fwr4_aa', 'cdr1_aa', 'cdr2_aa', 'cdr3_aa', 'sequence_alignment_aa', 'v_sequence_alignment_aa', 'd_sequence_alignment_aa', 'j_sequence_alignment_aa', 'complete_vdj', 'j_call_multimappers', 'j_call_multiplicity', 'j_call_sequence_start_multimappers', 'j_call_sequence_end_multimappers', 'j_call_support_multimappers', 'mu_count', 'ambiguous', 'rearrangement_status', 'clone_id', 'changeo_clone_id'
    metadata: 'clone_id', 'clone_id_by_size', 'sample_id', 'locus_VDJ', 'locus_VJ', 'productive_VDJ', 'productive_VJ', 'v_call_genotyped_VDJ', 'd_call_VDJ', 'j_call_VDJ', 'v_call_genotyped_VJ', 'j_call_VJ', 'c_call_VDJ', 'c_call_VJ', 'junction_VDJ', 'junction_VJ', 'junction_aa_VDJ', 'junction_aa_VJ', 'v_call_genotyped_B_VDJ', 'd_call_B_VDJ', 'j_call_B_VDJ', 'v_call_genotyped_B_VJ', 'j_call_B_VJ', 'c_call_B_VDJ', 'c_call_B_VJ', 'productive_B_VDJ', 'productive_B_VJ', 'duplicate_count_B_VDJ', 'duplicate_count_B_VJ', 'v_call_VDJ_main', 'v_call_VJ_main', 'd_call_VDJ_main', 'j_call_VDJ_main', 'j_call_VJ_main', 'c_call_VDJ_main', 'c_call_VJ_main', 'v_call_B_VDJ_main', 'd_call_B_VDJ_main', 'j_call_B_VDJ_main', 'v_call_B_VJ_main', 'j_call_B_VJ_main', 'isotype', 'isotype_status', 'locus_status', 'chain_status', 'rearrangement_status_VDJ', 'rearrangement_status_VJ', 'changeo_clone_id'
    layout: layout for 1 vertices, layout for 0 vertices
    graph: networkx graph of 1 vertices, networkx graph of 0 vertices

copy

You can deep copy the Dandelion object to another variable which will inherit all slots:

[8]:

vdj2 = vdj.copy()
vdj2.metadata

[8]:

	clone_id	clone_id_by_size	sample_id	locus_VDJ	locus_VJ	productive_VDJ	productive_VJ	v_call_genotyped_VDJ	d_call_VDJ	j_call_VDJ	...	j_call_B_VDJ_main	v_call_B_VJ_main	j_call_B_VJ_main	isotype	isotype_status	locus_status	chain_status	rearrangement_status_VDJ	rearrangement_status_VJ	changeo_clone_id
sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC	B_VDJ_119_3_2_VJ_80_2_3	1952	sc5p_v2_hs_PBMC_10k	IGH	IGK	T	T	IGHV1-69,IGHV1-69D	IGHD3-22	IGHJ3	...	IGHJ3	IGKV1-8	IGKJ1	IgM	IgM	IGH + IGK	Single pair	standard	standard	11_0
sc5p_v2_hs_PBMC_10k_AAACCTGTCGAGAACG	B_VDJ_42_1_2_VJ_54_1_1	1567	sc5p_v2_hs_PBMC_10k	IGH	IGL	T	T	IGHV1-2	None	IGHJ3	...	IGHJ3	IGLV5-45	IGLJ3	IgM	IgM	IGH + IGL	Single pair	standard	standard	150_1
sc5p_v2_hs_PBMC_10k_AAACCTGTCTTGAGAC	B_VDJ_38_4_4_VJ_191_1_1	1568	sc5p_v2_hs_PBMC_10k	IGH	IGK	T	T	IGHV5-51	None	IGHJ3	...	IGHJ3	IGKV1D-8	IGKJ2	IgM	IgM	IGH + IGK	Single pair	standard	standard	322_2
sc5p_v2_hs_PBMC_10k_AAACGGGAGCGACGTA	B_VDJ_55_2_1_VJ_184_2_7	1569	sc5p_v2_hs_PBMC_10k	IGH	IGL	T	T	IGHV4-4	IGHD6-13	IGHJ3	...	IGHJ3	IGLV3-19	IGLJ3,IGLJ2	IgM	IgM	IGH + IGL	Single pair	standard	standard	290_3
sc5p_v2_hs_PBMC_10k_AAACGGGCACTGTTAG	B_VDJ_44_2_3_VJ_164_3_4	1570	sc5p_v2_hs_PBMC_10k	IGH	IGL	T	T	IGHV4-39	IGHD3-22	IGHJ3	...	IGHJ3	IGLV3-21	IGLJ3,IGLJ2	IgM	IgM	IGH + IGL	Single pair	standard	standard	518_4
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
vdj_v1_hs_pbmc3_TTTCCTCAGCAATATG	B_VDJ_41_2_1_VJ_26_2_8	796	vdj_v1_hs_pbmc3	IGH	IGK	T	T	IGHV2-5	IGHD5/OR15-5a,IGHD5/OR15-5b	IGHJ5,IGHJ4	...	IGHJ5,IGHJ4	IGKV4-1	IGKJ4	IgM	IgM	IGH + IGK	Single pair	standard	standard	1762_1974
vdj_v1_hs_pbmc3_TTTCCTCAGCGCTTAT	B_VDJ_2_6_3_VJ_87_1_3	797	vdj_v1_hs_pbmc3	IGH	IGK	T	T	IGHV3-30	IGHD4-17	IGHJ6	...	IGHJ6	IGKV2-30	IGKJ2	IgM	IgM	IGH + IGK	Single pair	standard	standard	833_1975
vdj_v1_hs_pbmc3_TTTCCTCAGGGAAACA	B_VDJ_1_1_1_VJ_139_4_16	798	vdj_v1_hs_pbmc3	IGH	IGK	T	T	IGHV4-59	IGHD6-13	IGHJ2	...	IGHJ2	IGKV1D-39,IGKV1-39	IGKJ1	IgM	IgM	IGH + IGK	Single pair	standard	standard	1203_1976
vdj_v1_hs_pbmc3_TTTGCGCCATACCATG	B_VDJ_47_4_1_VJ_103_3_4	799	vdj_v1_hs_pbmc3	IGH	IGL	T	T	IGHV1-69,IGHV1-69D	IGHD2-15	IGHJ6	...	IGHJ6	IGLV1-47	IGLJ3	IgM	IgM	IGH + IGL	Single pair	standard	standard	1803_1977
vdj_v1_hs_pbmc3_TTTGGTTGTAGGCATG	B_VDJ_184_5_1_VJ_121_3_3	2336	vdj_v1_hs_pbmc3	IGH	IGL	T	T	IGHV3-23,IGHV3-23D	None	IGHJ4	...	IGHJ4	IGLV2-11	IGLJ3,IGLJ2	IgM	IgM	IGH + IGL	Single pair	standard	standard	1937_1978

2071 rows × 48 columns

Retrieving entries with `update_metadata`

The .metadata slot in Dandelion class automatically initializes whenever the .data slot is filled. However, it only returns a standard number of columns that are pre-specified. To retrieve other columns from the .data slot, we can update the metadata with ddl.update_metadata and specify the options retrieve and retrieve_mode.

The following modes determine how the retrieval is completed:

split and unique only - splits the retrieval into VDJ and VJ chains. A | will separate unique element.

split and merge - splits the retrieval into VDJ and VJ chains. A | will separate every element.

merge and unique only - smiliar to above but merged into a single column.

split - split retrieval into individual columns for each contig.

merge - merge retrieval into a single column where a | will separate every element.

For numerical columns, there’s additional options:

split and sum - splits the retrieval into VDJ and VJ chains and sum separately.

split and average - smiliar to above but average instead of sum.

sum - sum the retrievals into a single column.

average - averages the retrievals into a single column.

If retrieve_mode is not specified, it will default to split and merge

Example: retrieving fwr1 sequences

[9]:

ddl.update_metadata(vdj, retrieve="fwr1")
vdj

[9]:

Dandelion class object with n_obs = 2071 and n_contigs = 4882
    data: 'sequence_id', 'sequence', 'rev_comp', 'productive', 'v_call', 'd_call', 'j_call', 'sequence_alignment', 'germline_alignment', 'junction', 'junction_aa', 'v_cigar', 'd_cigar', 'j_cigar', 'stop_codon', 'vj_in_frame', 'locus', 'c_call', 'junction_length', 'np1_length', 'np2_length', 'v_sequence_start', 'v_sequence_end', 'v_germline_start', 'v_germline_end', 'd_sequence_start', 'd_sequence_end', 'd_germline_start', 'd_germline_end', 'j_sequence_start', 'j_sequence_end', 'j_germline_start', 'j_germline_end', 'v_score', 'v_identity', 'v_support', 'd_score', 'd_identity', 'd_support', 'j_score', 'j_identity', 'j_support', 'fwr1', 'fwr2', 'fwr3', 'fwr4', 'cdr1', 'cdr2', 'cdr3', 'cell_id', 'consensus_count', 'duplicate_count', 'v_call_10x', 'd_call_10x', 'j_call_10x', 'junction_10x', 'junction_10x_aa', 'j_support_igblastn', 'j_score_igblastn', 'j_call_igblastn', 'j_call_blastn', 'j_identity_blastn', 'j_alignment_length_blastn', 'j_number_of_mismatches_blastn', 'j_number_of_gap_openings_blastn', 'j_sequence_start_blastn', 'j_sequence_end_blastn', 'j_germline_start_blastn', 'j_germline_end_blastn', 'j_support_blastn', 'j_score_blastn', 'j_sequence_alignment_blastn', 'j_germline_alignment_blastn', 'j_source', 'd_support_igblastn', 'd_score_igblastn', 'd_call_igblastn', 'd_call_blastn', 'd_identity_blastn', 'd_alignment_length_blastn', 'd_number_of_mismatches_blastn', 'd_number_of_gap_openings_blastn', 'd_sequence_start_blastn', 'd_sequence_end_blastn', 'd_germline_start_blastn', 'd_germline_end_blastn', 'd_support_blastn', 'd_score_blastn', 'd_sequence_alignment_blastn', 'd_germline_alignment_blastn', 'd_source', 'v_call_genotyped', 'germline_alignment_d_mask', 'sample_id', 'c_sequence_alignment', 'c_germline_alignment', 'c_sequence_start', 'c_sequence_end', 'c_score', 'c_identity', 'c_call_10x', 'junction_aa_length', 'fwr1_aa', 'fwr2_aa', 'fwr3_aa', 'fwr4_aa', 'cdr1_aa', 'cdr2_aa', 'cdr3_aa', 'sequence_alignment_aa', 'v_sequence_alignment_aa', 'd_sequence_alignment_aa', 'j_sequence_alignment_aa', 'complete_vdj', 'j_call_multimappers', 'j_call_multiplicity', 'j_call_sequence_start_multimappers', 'j_call_sequence_end_multimappers', 'j_call_support_multimappers', 'mu_count', 'ambiguous', 'rearrangement_status', 'clone_id', 'changeo_clone_id'
    metadata: 'clone_id', 'clone_id_by_size', 'sample_id', 'locus_VDJ', 'locus_VJ', 'productive_VDJ', 'productive_VJ', 'v_call_genotyped_VDJ', 'd_call_VDJ', 'j_call_VDJ', 'v_call_genotyped_VJ', 'j_call_VJ', 'c_call_VDJ', 'c_call_VJ', 'junction_VDJ', 'junction_VJ', 'junction_aa_VDJ', 'junction_aa_VJ', 'v_call_genotyped_B_VDJ', 'd_call_B_VDJ', 'j_call_B_VDJ', 'v_call_genotyped_B_VJ', 'j_call_B_VJ', 'c_call_B_VDJ', 'c_call_B_VJ', 'productive_B_VDJ', 'productive_B_VJ', 'duplicate_count_B_VDJ', 'duplicate_count_B_VJ', 'v_call_VDJ_main', 'v_call_VJ_main', 'd_call_VDJ_main', 'j_call_VDJ_main', 'j_call_VJ_main', 'c_call_VDJ_main', 'c_call_VJ_main', 'v_call_B_VDJ_main', 'd_call_B_VDJ_main', 'j_call_B_VDJ_main', 'v_call_B_VJ_main', 'j_call_B_VJ_main', 'isotype', 'isotype_status', 'locus_status', 'chain_status', 'rearrangement_status_VDJ', 'rearrangement_status_VJ', 'changeo_clone_id', 'fwr1_VDJ', 'fwr1_VJ'
    layout: layout for 2071 vertices, layout for 70 vertices
    graph: networkx graph of 2071 vertices, networkx graph of 70 vertices

Note the additional fwr1 VDJ and VJ columns in the metadata slot.

By default, dandelion will not try to merge numerical columns as it can create mixed dtype columns.

There is a new sub-function that will try and retrieve frequently used columns such as np1_length, np2_length:

[10]:

vdj.update_plus()
vdj

/opt/homebrew/Caskroom/mambaforge/base/envs/dandelion/lib/python3.11/site-packages/numpy/core/fromnumeric.py:3464: RuntimeWarning: Mean of empty slice.
/opt/homebrew/Caskroom/mambaforge/base/envs/dandelion/lib/python3.11/site-packages/numpy/core/_methods.py:192: RuntimeWarning: invalid value encountered in scalar divide

[10]:

Dandelion class object with n_obs = 2071 and n_contigs = 4882
    data: 'sequence_id', 'sequence', 'rev_comp', 'productive', 'v_call', 'd_call', 'j_call', 'sequence_alignment', 'germline_alignment', 'junction', 'junction_aa', 'v_cigar', 'd_cigar', 'j_cigar', 'stop_codon', 'vj_in_frame', 'locus', 'c_call', 'junction_length', 'np1_length', 'np2_length', 'v_sequence_start', 'v_sequence_end', 'v_germline_start', 'v_germline_end', 'd_sequence_start', 'd_sequence_end', 'd_germline_start', 'd_germline_end', 'j_sequence_start', 'j_sequence_end', 'j_germline_start', 'j_germline_end', 'v_score', 'v_identity', 'v_support', 'd_score', 'd_identity', 'd_support', 'j_score', 'j_identity', 'j_support', 'fwr1', 'fwr2', 'fwr3', 'fwr4', 'cdr1', 'cdr2', 'cdr3', 'cell_id', 'consensus_count', 'duplicate_count', 'v_call_10x', 'd_call_10x', 'j_call_10x', 'junction_10x', 'junction_10x_aa', 'j_support_igblastn', 'j_score_igblastn', 'j_call_igblastn', 'j_call_blastn', 'j_identity_blastn', 'j_alignment_length_blastn', 'j_number_of_mismatches_blastn', 'j_number_of_gap_openings_blastn', 'j_sequence_start_blastn', 'j_sequence_end_blastn', 'j_germline_start_blastn', 'j_germline_end_blastn', 'j_support_blastn', 'j_score_blastn', 'j_sequence_alignment_blastn', 'j_germline_alignment_blastn', 'j_source', 'd_support_igblastn', 'd_score_igblastn', 'd_call_igblastn', 'd_call_blastn', 'd_identity_blastn', 'd_alignment_length_blastn', 'd_number_of_mismatches_blastn', 'd_number_of_gap_openings_blastn', 'd_sequence_start_blastn', 'd_sequence_end_blastn', 'd_germline_start_blastn', 'd_germline_end_blastn', 'd_support_blastn', 'd_score_blastn', 'd_sequence_alignment_blastn', 'd_germline_alignment_blastn', 'd_source', 'v_call_genotyped', 'germline_alignment_d_mask', 'sample_id', 'c_sequence_alignment', 'c_germline_alignment', 'c_sequence_start', 'c_sequence_end', 'c_score', 'c_identity', 'c_call_10x', 'junction_aa_length', 'fwr1_aa', 'fwr2_aa', 'fwr3_aa', 'fwr4_aa', 'cdr1_aa', 'cdr2_aa', 'cdr3_aa', 'sequence_alignment_aa', 'v_sequence_alignment_aa', 'd_sequence_alignment_aa', 'j_sequence_alignment_aa', 'complete_vdj', 'j_call_multimappers', 'j_call_multiplicity', 'j_call_sequence_start_multimappers', 'j_call_sequence_end_multimappers', 'j_call_support_multimappers', 'mu_count', 'ambiguous', 'rearrangement_status', 'clone_id', 'changeo_clone_id'
    metadata: 'clone_id', 'clone_id_by_size', 'sample_id', 'locus_VDJ', 'locus_VJ', 'productive_VDJ', 'productive_VJ', 'v_call_genotyped_VDJ', 'd_call_VDJ', 'j_call_VDJ', 'v_call_genotyped_VJ', 'j_call_VJ', 'c_call_VDJ', 'c_call_VJ', 'junction_VDJ', 'junction_VJ', 'junction_aa_VDJ', 'junction_aa_VJ', 'v_call_genotyped_B_VDJ', 'd_call_B_VDJ', 'j_call_B_VDJ', 'v_call_genotyped_B_VJ', 'j_call_B_VJ', 'c_call_B_VDJ', 'c_call_B_VJ', 'productive_B_VDJ', 'productive_B_VJ', 'duplicate_count_B_VDJ', 'duplicate_count_B_VJ', 'v_call_VDJ_main', 'v_call_VJ_main', 'd_call_VDJ_main', 'j_call_VDJ_main', 'j_call_VJ_main', 'c_call_VDJ_main', 'c_call_VJ_main', 'v_call_B_VDJ_main', 'd_call_B_VDJ_main', 'j_call_B_VDJ_main', 'v_call_B_VJ_main', 'j_call_B_VJ_main', 'isotype', 'isotype_status', 'locus_status', 'chain_status', 'rearrangement_status_VDJ', 'rearrangement_status_VJ', 'changeo_clone_id', 'fwr1_VDJ', 'fwr1_VJ', 'mu_count_VDJ', 'mu_count_VJ', 'mu_count', 'junction_length_VDJ', 'junction_length_VJ', 'junction_aa_length_VDJ', 'junction_aa_length_VJ', 'np1_length_VDJ', 'np1_length_VJ', 'np2_length_VDJ'
    layout: layout for 2071 vertices, layout for 70 vertices
    graph: networkx graph of 2071 vertices, networkx graph of 70 vertices

concatenating multiple objects

This is a simple function to concatenate (append) two or more Dandelion class, or pandas dataframes. Note that this operates on the .data slot and not the .metadata slot.

[12]:

# for example, the original dandelion class has 2071 unique cell barcodes and 4882 contigs
vdj

[12]:

Dandelion class object with n_obs = 2071 and n_contigs = 4882
    data: 'sequence_id', 'sequence', 'rev_comp', 'productive', 'v_call', 'd_call', 'j_call', 'sequence_alignment', 'germline_alignment', 'junction', 'junction_aa', 'v_cigar', 'd_cigar', 'j_cigar', 'stop_codon', 'vj_in_frame', 'locus', 'c_call', 'junction_length', 'np1_length', 'np2_length', 'v_sequence_start', 'v_sequence_end', 'v_germline_start', 'v_germline_end', 'd_sequence_start', 'd_sequence_end', 'd_germline_start', 'd_germline_end', 'j_sequence_start', 'j_sequence_end', 'j_germline_start', 'j_germline_end', 'v_score', 'v_identity', 'v_support', 'd_score', 'd_identity', 'd_support', 'j_score', 'j_identity', 'j_support', 'fwr1', 'fwr2', 'fwr3', 'fwr4', 'cdr1', 'cdr2', 'cdr3', 'cell_id', 'consensus_count', 'duplicate_count', 'v_call_10x', 'd_call_10x', 'j_call_10x', 'junction_10x', 'junction_10x_aa', 'j_support_igblastn', 'j_score_igblastn', 'j_call_igblastn', 'j_call_blastn', 'j_identity_blastn', 'j_alignment_length_blastn', 'j_number_of_mismatches_blastn', 'j_number_of_gap_openings_blastn', 'j_sequence_start_blastn', 'j_sequence_end_blastn', 'j_germline_start_blastn', 'j_germline_end_blastn', 'j_support_blastn', 'j_score_blastn', 'j_sequence_alignment_blastn', 'j_germline_alignment_blastn', 'j_source', 'd_support_igblastn', 'd_score_igblastn', 'd_call_igblastn', 'd_call_blastn', 'd_identity_blastn', 'd_alignment_length_blastn', 'd_number_of_mismatches_blastn', 'd_number_of_gap_openings_blastn', 'd_sequence_start_blastn', 'd_sequence_end_blastn', 'd_germline_start_blastn', 'd_germline_end_blastn', 'd_support_blastn', 'd_score_blastn', 'd_sequence_alignment_blastn', 'd_germline_alignment_blastn', 'd_source', 'v_call_genotyped', 'germline_alignment_d_mask', 'sample_id', 'c_sequence_alignment', 'c_germline_alignment', 'c_sequence_start', 'c_sequence_end', 'c_score', 'c_identity', 'c_call_10x', 'junction_aa_length', 'fwr1_aa', 'fwr2_aa', 'fwr3_aa', 'fwr4_aa', 'cdr1_aa', 'cdr2_aa', 'cdr3_aa', 'sequence_alignment_aa', 'v_sequence_alignment_aa', 'd_sequence_alignment_aa', 'j_sequence_alignment_aa', 'complete_vdj', 'j_call_multimappers', 'j_call_multiplicity', 'j_call_sequence_start_multimappers', 'j_call_sequence_end_multimappers', 'j_call_support_multimappers', 'mu_count', 'ambiguous', 'rearrangement_status', 'clone_id', 'changeo_clone_id'
    metadata: 'clone_id', 'clone_id_by_size', 'sample_id', 'locus_VDJ', 'locus_VJ', 'productive_VDJ', 'productive_VJ', 'v_call_genotyped_VDJ', 'd_call_VDJ', 'j_call_VDJ', 'v_call_genotyped_VJ', 'j_call_VJ', 'c_call_VDJ', 'c_call_VJ', 'junction_VDJ', 'junction_VJ', 'junction_aa_VDJ', 'junction_aa_VJ', 'v_call_genotyped_B_VDJ', 'd_call_B_VDJ', 'j_call_B_VDJ', 'v_call_genotyped_B_VJ', 'j_call_B_VJ', 'c_call_B_VDJ', 'c_call_B_VJ', 'productive_B_VDJ', 'productive_B_VJ', 'duplicate_count_B_VDJ', 'duplicate_count_B_VJ', 'v_call_VDJ_main', 'v_call_VJ_main', 'd_call_VDJ_main', 'j_call_VDJ_main', 'j_call_VJ_main', 'c_call_VDJ_main', 'c_call_VJ_main', 'v_call_B_VDJ_main', 'd_call_B_VDJ_main', 'j_call_B_VDJ_main', 'v_call_B_VJ_main', 'j_call_B_VJ_main', 'isotype', 'isotype_status', 'locus_status', 'chain_status', 'rearrangement_status_VDJ', 'rearrangement_status_VJ', 'changeo_clone_id', 'fwr1_VDJ', 'fwr1_VJ', 'mu_count_VDJ', 'mu_count_VJ', 'mu_count', 'junction_length_VDJ', 'junction_length_VJ', 'junction_aa_length_VDJ', 'junction_aa_length_VJ', 'np1_length_VDJ', 'np1_length_VJ', 'np2_length_VDJ'
    layout: layout for 2071 vertices, layout for 70 vertices
    graph: networkx graph of 2071 vertices, networkx graph of 70 vertices

[13]:

# now it has 14646 (4882*3) contigs instead, and the metadata should also be properly populated
vdj_concat = ddl.concat([vdj, vdj, vdj])
vdj_concat

[13]:

Dandelion class object with n_obs = 2071 and n_contigs = 14646
    data: 'sequence_id', 'sequence', 'rev_comp', 'productive', 'v_call', 'd_call', 'j_call', 'sequence_alignment', 'germline_alignment', 'junction', 'junction_aa', 'v_cigar', 'd_cigar', 'j_cigar', 'stop_codon', 'vj_in_frame', 'locus', 'c_call', 'junction_length', 'np1_length', 'np2_length', 'v_sequence_start', 'v_sequence_end', 'v_germline_start', 'v_germline_end', 'd_sequence_start', 'd_sequence_end', 'd_germline_start', 'd_germline_end', 'j_sequence_start', 'j_sequence_end', 'j_germline_start', 'j_germline_end', 'v_score', 'v_identity', 'v_support', 'd_score', 'd_identity', 'd_support', 'j_score', 'j_identity', 'j_support', 'fwr1', 'fwr2', 'fwr3', 'fwr4', 'cdr1', 'cdr2', 'cdr3', 'cell_id', 'consensus_count', 'duplicate_count', 'v_call_10x', 'd_call_10x', 'j_call_10x', 'junction_10x', 'junction_10x_aa', 'j_support_igblastn', 'j_score_igblastn', 'j_call_igblastn', 'j_call_blastn', 'j_identity_blastn', 'j_alignment_length_blastn', 'j_number_of_mismatches_blastn', 'j_number_of_gap_openings_blastn', 'j_sequence_start_blastn', 'j_sequence_end_blastn', 'j_germline_start_blastn', 'j_germline_end_blastn', 'j_support_blastn', 'j_score_blastn', 'j_sequence_alignment_blastn', 'j_germline_alignment_blastn', 'j_source', 'd_support_igblastn', 'd_score_igblastn', 'd_call_igblastn', 'd_call_blastn', 'd_identity_blastn', 'd_alignment_length_blastn', 'd_number_of_mismatches_blastn', 'd_number_of_gap_openings_blastn', 'd_sequence_start_blastn', 'd_sequence_end_blastn', 'd_germline_start_blastn', 'd_germline_end_blastn', 'd_support_blastn', 'd_score_blastn', 'd_sequence_alignment_blastn', 'd_germline_alignment_blastn', 'd_source', 'v_call_genotyped', 'germline_alignment_d_mask', 'sample_id', 'c_sequence_alignment', 'c_germline_alignment', 'c_sequence_start', 'c_sequence_end', 'c_score', 'c_identity', 'c_call_10x', 'junction_aa_length', 'fwr1_aa', 'fwr2_aa', 'fwr3_aa', 'fwr4_aa', 'cdr1_aa', 'cdr2_aa', 'cdr3_aa', 'sequence_alignment_aa', 'v_sequence_alignment_aa', 'd_sequence_alignment_aa', 'j_sequence_alignment_aa', 'complete_vdj', 'j_call_multimappers', 'j_call_multiplicity', 'j_call_sequence_start_multimappers', 'j_call_sequence_end_multimappers', 'j_call_support_multimappers', 'mu_count', 'ambiguous', 'rearrangement_status', 'clone_id', 'changeo_clone_id'
    metadata: 'clone_id', 'clone_id_by_size', 'sample_id', 'locus_VDJ', 'locus_VJ', 'productive_VDJ', 'productive_VJ', 'v_call_genotyped_VDJ', 'd_call_VDJ', 'j_call_VDJ', 'v_call_genotyped_VJ', 'j_call_VJ', 'c_call_VDJ', 'c_call_VJ', 'junction_VDJ', 'junction_VJ', 'junction_aa_VDJ', 'junction_aa_VJ', 'v_call_genotyped_B_VDJ', 'd_call_B_VDJ', 'j_call_B_VDJ', 'v_call_genotyped_B_VJ', 'j_call_B_VJ', 'c_call_B_VDJ', 'c_call_B_VJ', 'productive_B_VDJ', 'productive_B_VJ', 'duplicate_count_B_VDJ', 'duplicate_count_B_VJ', 'v_call_VDJ_main', 'v_call_VJ_main', 'd_call_VDJ_main', 'j_call_VDJ_main', 'j_call_VJ_main', 'c_call_VDJ_main', 'c_call_VJ_main', 'v_call_B_VDJ_main', 'd_call_B_VDJ_main', 'j_call_B_VDJ_main', 'v_call_B_VJ_main', 'j_call_B_VJ_main', 'isotype', 'isotype_status', 'locus_status', 'chain_status', 'rearrangement_status_VDJ', 'rearrangement_status_VJ'

[14]:

vdj_concat.data[["sequence_id", "cell_id"]].head()

[14]:

	sequence_id	cell_id
sequence_id
sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC_contig_2-0	sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC_contig_2-0	sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC
sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC_contig_2-1	sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC_contig_2-1	sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC
sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC_contig_2-2	sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC_contig_2-2	sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC
sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC_contig_1-0	sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC_contig_1-0	sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC
sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC_contig_1-1	sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC_contig_1-1	sc5p_v2_hs_PBMC_10k_AAACCTGTCCGTTGTC

ddl.concat also lets you add in your custom prefixes/suffixes to append to the sequence ids. If not provided, it will add -0, -1 etc. as a suffix if it detects that the sequence ids are not unique as seen above.

read/write

Dandelion class can be saved using .write_h5ddl and .write_pkl functions with accompanying compression methods. write_h5ddl primarily uses pandas to_hdf library and write_pkl just uses pickle. read_h5ddl and read_pkl functions will read the respective file formats accordingly.

[15]:

%time vdj.write_h5ddl('dandelion_results.h5ddl', complib = 'bzip2')

/Users/uqztuong/Library/CloudStorage/OneDrive-TheUniversityofQueensland/Documents/GitHub/dandelion/dandelion/utilities/_core.py:1120: PerformanceWarning:
your performance may suffer as PyTables will pickle object types that it cannot
map directly to c-types [inferred_type->mixed-integer,key->block2_values] [items->Index(['sequence_id', 'sequence', 'rev_comp', 'productive', 'v_call', 'd_call',
       'j_call', 'sequence_alignment', 'germline_alignment', 'junction',
       'junction_aa', 'v_cigar', 'd_cigar', 'j_cigar', 'stop_codon',
       'vj_in_frame', 'locus', 'c_call', 'fwr1', 'fwr2', 'fwr3', 'fwr4',
       'cdr1', 'cdr2', 'cdr3', 'cell_id', 'v_call_10x', 'd_call_10x',
       'j_call_10x', 'junction_10x', 'junction_10x_aa', 'j_call_igblastn',
       'j_call_blastn', 'j_sequence_alignment_blastn',
       'j_germline_alignment_blastn', 'd_call_igblastn', 'd_call_blastn',
       'd_sequence_alignment_blastn', 'd_germline_alignment_blastn',
       'd_source', 'v_call_genotyped', 'germline_alignment_d_mask',
       'sample_id', 'c_sequence_alignment', 'c_germline_alignment',
       'c_call_10x', 'fwr1_aa', 'fwr2_aa', 'fwr3_aa', 'fwr4_aa', 'cdr1_aa',
       'cdr2_aa', 'cdr3_aa', 'sequence_alignment_aa',
       'v_sequence_alignment_aa', 'd_sequence_alignment_aa',
       'j_sequence_alignment_aa', 'complete_vdj', 'j_call_multimappers',
       'j_call_sequence_start_multimappers',
       'j_call_sequence_end_multimappers', 'j_call_support_multimappers',
       'ambiguous', 'rearrangement_status', 'clone_id', 'changeo_clone_id'],
      dtype='object')]

CPU times: user 1.74 s, sys: 94.3 ms, total: 1.84 s
Wall time: 1.93 s

If you see any warnings above, it’s due to mix dtypes somewhere in the object. So do some checking if you think it will interfere with downstream usage.

[16]:

%time vdj_1 = ddl.read_h5ddl('dandelion_results.h5ddl')
vdj_1

CPU times: user 568 ms, sys: 50.8 ms, total: 618 ms
Wall time: 603 ms

[16]:

Dandelion class object with n_obs = 2071 and n_contigs = 4882
    data: 'sequence_id', 'sequence', 'rev_comp', 'productive', 'v_call', 'd_call', 'j_call', 'sequence_alignment', 'germline_alignment', 'junction', 'junction_aa', 'v_cigar', 'd_cigar', 'j_cigar', 'stop_codon', 'vj_in_frame', 'locus', 'c_call', 'junction_length', 'np1_length', 'np2_length', 'v_sequence_start', 'v_sequence_end', 'v_germline_start', 'v_germline_end', 'd_sequence_start', 'd_sequence_end', 'd_germline_start', 'd_germline_end', 'j_sequence_start', 'j_sequence_end', 'j_germline_start', 'j_germline_end', 'v_score', 'v_identity', 'v_support', 'd_score', 'd_identity', 'd_support', 'j_score', 'j_identity', 'j_support', 'fwr1', 'fwr2', 'fwr3', 'fwr4', 'cdr1', 'cdr2', 'cdr3', 'cell_id', 'consensus_count', 'duplicate_count', 'v_call_10x', 'd_call_10x', 'j_call_10x', 'junction_10x', 'junction_10x_aa', 'j_support_igblastn', 'j_score_igblastn', 'j_call_igblastn', 'j_call_blastn', 'j_identity_blastn', 'j_alignment_length_blastn', 'j_number_of_mismatches_blastn', 'j_number_of_gap_openings_blastn', 'j_sequence_start_blastn', 'j_sequence_end_blastn', 'j_germline_start_blastn', 'j_germline_end_blastn', 'j_support_blastn', 'j_score_blastn', 'j_sequence_alignment_blastn', 'j_germline_alignment_blastn', 'j_source', 'd_support_igblastn', 'd_score_igblastn', 'd_call_igblastn', 'd_call_blastn', 'd_identity_blastn', 'd_alignment_length_blastn', 'd_number_of_mismatches_blastn', 'd_number_of_gap_openings_blastn', 'd_sequence_start_blastn', 'd_sequence_end_blastn', 'd_germline_start_blastn', 'd_germline_end_blastn', 'd_support_blastn', 'd_score_blastn', 'd_sequence_alignment_blastn', 'd_germline_alignment_blastn', 'd_source', 'v_call_genotyped', 'germline_alignment_d_mask', 'sample_id', 'c_sequence_alignment', 'c_germline_alignment', 'c_sequence_start', 'c_sequence_end', 'c_score', 'c_identity', 'c_call_10x', 'junction_aa_length', 'fwr1_aa', 'fwr2_aa', 'fwr3_aa', 'fwr4_aa', 'cdr1_aa', 'cdr2_aa', 'cdr3_aa', 'sequence_alignment_aa', 'v_sequence_alignment_aa', 'd_sequence_alignment_aa', 'j_sequence_alignment_aa', 'complete_vdj', 'j_call_multimappers', 'j_call_multiplicity', 'j_call_sequence_start_multimappers', 'j_call_sequence_end_multimappers', 'j_call_support_multimappers', 'mu_count', 'ambiguous', 'rearrangement_status', 'clone_id', 'changeo_clone_id'
    metadata: 'clone_id', 'clone_id_by_size', 'sample_id', 'locus_VDJ', 'locus_VJ', 'productive_VDJ', 'productive_VJ', 'v_call_genotyped_VDJ', 'd_call_VDJ', 'j_call_VDJ', 'v_call_genotyped_VJ', 'j_call_VJ', 'c_call_VDJ', 'c_call_VJ', 'junction_VDJ', 'junction_VJ', 'junction_aa_VDJ', 'junction_aa_VJ', 'v_call_genotyped_B_VDJ', 'd_call_B_VDJ', 'j_call_B_VDJ', 'v_call_genotyped_B_VJ', 'j_call_B_VJ', 'c_call_B_VDJ', 'c_call_B_VJ', 'productive_B_VDJ', 'productive_B_VJ', 'duplicate_count_B_VDJ', 'duplicate_count_B_VJ', 'v_call_VDJ_main', 'v_call_VJ_main', 'd_call_VDJ_main', 'j_call_VDJ_main', 'j_call_VJ_main', 'c_call_VDJ_main', 'c_call_VJ_main', 'v_call_B_VDJ_main', 'd_call_B_VDJ_main', 'j_call_B_VDJ_main', 'v_call_B_VJ_main', 'j_call_B_VJ_main', 'isotype', 'isotype_status', 'locus_status', 'chain_status', 'rearrangement_status_VDJ', 'rearrangement_status_VJ', 'changeo_clone_id', 'fwr1_VDJ', 'fwr1_VJ', 'mu_count_VDJ', 'mu_count_VJ', 'mu_count', 'junction_length_VDJ', 'junction_length_VJ', 'junction_aa_length_VDJ', 'junction_aa_length_VJ', 'np1_length_VDJ', 'np1_length_VJ', 'np2_length_VDJ'
    layout: layout for 2071 vertices, layout for 70 vertices
    graph: networkx graph of 2071 vertices, networkx graph of 70 vertices

The read/write times using pickle can be situationally faster/slower and file sizes can also be situationally smaller/larger (depending on which compression is used).

[17]:

%time vdj.write_pkl('dandelion_results.pkl.gz')

CPU times: user 5.43 s, sys: 10.8 ms, total: 5.44 s
Wall time: 5.47 s

[18]:

%time vdj_2 = ddl.read_pkl('dandelion_results.pkl.gz')
vdj_2

CPU times: user 181 ms, sys: 14.5 ms, total: 196 ms
Wall time: 200 ms

[18]:

Dandelion class object with n_obs = 2071 and n_contigs = 4882
    data: 'sequence_id', 'sequence', 'rev_comp', 'productive', 'v_call', 'd_call', 'j_call', 'sequence_alignment', 'germline_alignment', 'junction', 'junction_aa', 'v_cigar', 'd_cigar', 'j_cigar', 'stop_codon', 'vj_in_frame', 'locus', 'c_call', 'junction_length', 'np1_length', 'np2_length', 'v_sequence_start', 'v_sequence_end', 'v_germline_start', 'v_germline_end', 'd_sequence_start', 'd_sequence_end', 'd_germline_start', 'd_germline_end', 'j_sequence_start', 'j_sequence_end', 'j_germline_start', 'j_germline_end', 'v_score', 'v_identity', 'v_support', 'd_score', 'd_identity', 'd_support', 'j_score', 'j_identity', 'j_support', 'fwr1', 'fwr2', 'fwr3', 'fwr4', 'cdr1', 'cdr2', 'cdr3', 'cell_id', 'consensus_count', 'duplicate_count', 'v_call_10x', 'd_call_10x', 'j_call_10x', 'junction_10x', 'junction_10x_aa', 'j_support_igblastn', 'j_score_igblastn', 'j_call_igblastn', 'j_call_blastn', 'j_identity_blastn', 'j_alignment_length_blastn', 'j_number_of_mismatches_blastn', 'j_number_of_gap_openings_blastn', 'j_sequence_start_blastn', 'j_sequence_end_blastn', 'j_germline_start_blastn', 'j_germline_end_blastn', 'j_support_blastn', 'j_score_blastn', 'j_sequence_alignment_blastn', 'j_germline_alignment_blastn', 'j_source', 'd_support_igblastn', 'd_score_igblastn', 'd_call_igblastn', 'd_call_blastn', 'd_identity_blastn', 'd_alignment_length_blastn', 'd_number_of_mismatches_blastn', 'd_number_of_gap_openings_blastn', 'd_sequence_start_blastn', 'd_sequence_end_blastn', 'd_germline_start_blastn', 'd_germline_end_blastn', 'd_support_blastn', 'd_score_blastn', 'd_sequence_alignment_blastn', 'd_germline_alignment_blastn', 'd_source', 'v_call_genotyped', 'germline_alignment_d_mask', 'sample_id', 'c_sequence_alignment', 'c_germline_alignment', 'c_sequence_start', 'c_sequence_end', 'c_score', 'c_identity', 'c_call_10x', 'junction_aa_length', 'fwr1_aa', 'fwr2_aa', 'fwr3_aa', 'fwr4_aa', 'cdr1_aa', 'cdr2_aa', 'cdr3_aa', 'sequence_alignment_aa', 'v_sequence_alignment_aa', 'd_sequence_alignment_aa', 'j_sequence_alignment_aa', 'complete_vdj', 'j_call_multimappers', 'j_call_multiplicity', 'j_call_sequence_start_multimappers', 'j_call_sequence_end_multimappers', 'j_call_support_multimappers', 'mu_count', 'ambiguous', 'rearrangement_status', 'clone_id', 'changeo_clone_id'
    metadata: 'clone_id', 'clone_id_by_size', 'sample_id', 'locus_VDJ', 'locus_VJ', 'productive_VDJ', 'productive_VJ', 'v_call_genotyped_VDJ', 'd_call_VDJ', 'j_call_VDJ', 'v_call_genotyped_VJ', 'j_call_VJ', 'c_call_VDJ', 'c_call_VJ', 'junction_VDJ', 'junction_VJ', 'junction_aa_VDJ', 'junction_aa_VJ', 'v_call_genotyped_B_VDJ', 'd_call_B_VDJ', 'j_call_B_VDJ', 'v_call_genotyped_B_VJ', 'j_call_B_VJ', 'c_call_B_VDJ', 'c_call_B_VJ', 'productive_B_VDJ', 'productive_B_VJ', 'duplicate_count_B_VDJ', 'duplicate_count_B_VJ', 'v_call_VDJ_main', 'v_call_VJ_main', 'd_call_VDJ_main', 'j_call_VDJ_main', 'j_call_VJ_main', 'c_call_VDJ_main', 'c_call_VJ_main', 'v_call_B_VDJ_main', 'd_call_B_VDJ_main', 'j_call_B_VDJ_main', 'v_call_B_VJ_main', 'j_call_B_VJ_main', 'isotype', 'isotype_status', 'locus_status', 'chain_status', 'rearrangement_status_VDJ', 'rearrangement_status_VJ', 'changeo_clone_id', 'fwr1_VDJ', 'fwr1_VJ', 'mu_count_VDJ', 'mu_count_VJ', 'mu_count', 'junction_length_VDJ', 'junction_length_VJ', 'junction_aa_length_VDJ', 'junction_aa_length_VJ', 'np1_length_VDJ', 'np1_length_VJ', 'np2_length_VDJ'
    layout: layout for 2071 vertices, layout for 70 vertices
    graph: networkx graph of 2071 vertices, networkx graph of 70 vertices

[ ]: