diff --git a/data-pipeline/schemas/gnomad_v4_variants/prepare_gnomad_v4_exome_variants/input_path/mock_v4_release.ht.schema b/data-pipeline/schemas/gnomad_v4_variants/prepare_gnomad_v4_exome_variants/input_path/mock_v4_release.ht.schema
new file mode 100644
index 000000000..cb85362bc
--- /dev/null
+++ b/data-pipeline/schemas/gnomad_v4_variants/prepare_gnomad_v4_exome_variants/input_path/mock_v4_release.ht.schema
@@ -0,0 +1,338 @@
+----------------------------------------
+Global fields:
+    'freq_meta': array<dict<str, str>> 
+    'freq_index_dict': dict<str, int32> 
+    'faf_meta': array<dict<str, str>> 
+    'faf_index_dict': dict<str, int32> 
+    'freq_sample_count': array<int32> 
+    'filtering_model': struct {
+        model_name: str, 
+        score_name: str, 
+        feature_medians: dict<tuple (
+            str
+        ), struct {
+            a_index: int32, 
+            n_alt_alleles: int32, 
+            AS_pab_max: float64, 
+            AS_MQRankSum: float64, 
+            AS_SOR: float64, 
+            AS_ReadPosRankSum: float64
+        }>, 
+        variants_by_strata: dict<tuple (
+            str
+        ), int64>, 
+        features_importance: dict<str, float64>, 
+        features: array<str>, 
+        test_results: array<struct {
+            rf_prediction: str, 
+            rf_label: str, 
+            n: int32
+        }>, 
+        rf_snv_cutoff: struct {
+            bin: float64, 
+            min_score: float64
+        }, 
+        rf_indel_cutoff: struct {
+            bin: float64, 
+            min_score: float64
+        }, 
+        inbreeding_cutoff: float64, 
+        model_id: str
+    } 
+    'tool_versions': struct {
+        dbsnp_version: str, 
+        cadd_version: str, 
+        revel_version: str, 
+        splicaai_version: str, 
+        primateai_version: str, 
+        pangolin_version: str, 
+        vrs_version: str
+    } 
+    'vep_globals': struct {
+        vep_version: str, 
+        vep_csq_header: str, 
+        vep_help: str, 
+        vep_config: str
+    } 
+    'age_distribution': struct {
+        bin_edges: array<float64>, 
+        bin_freq: array<int32>, 
+        n_smaller: int32, 
+        n_larger: int32
+    } 
+    'age_index_dict': dict<str, int32> 
+    'age_meta': array<dict<str, str>> 
+    'grpmax_index_dict': dict<str, int32> 
+    'grpmax_meta': array<dict<str, str>> 
+    'README': dict<str, str> 
+    'gnomad_qc_repo': str 
+    'gnomad_methods_repo': str 
+----------------------------------------
+Row fields:
+    'locus': locus<GRCh38> 
+    'alleles': array<str> 
+    'freq': array<struct {
+        AC: int32, 
+        AF: float64, 
+        AN: int32, 
+        homozygote_count: int32
+    }> 
+    'grpmax': array<struct {
+        AC: int32, 
+        AF: float64, 
+        AN: int32, 
+        homozygote_count: int32, 
+        grp: str, 
+        faf95: float64
+    }> 
+    'faf': array<struct {
+        faf95: float64, 
+        faf99: float64
+    }> 
+    'a_index': int32 
+    'was_split': bool 
+    'rsid': set<str> 
+    'filters': set<str> 
+    'info': struct {
+        QUALapprox: int64, 
+        SB: array<int32>, 
+        MQ: float64, 
+        MQRankSum: float64, 
+        VarDP: int32, 
+        AS_ReadPosRankSum: float64, 
+        AS_pab_max: float64, 
+        AS_QD: float32, 
+        AS_MQ: float64, 
+        QD: float32, 
+        AS_MQRankSum: float64, 
+        FS: float64, 
+        AS_FS: float64, 
+        ReadPosRankSum: float64, 
+        AS_QUALapprox: int64, 
+        AS_SB_TABLE: array<int32>, 
+        AS_VarDP: int32, 
+        AS_SOR: float64, 
+        SOR: float64, 
+        singleton: bool, 
+        transmitted_singleton: bool, 
+        omni: bool, 
+        mills: bool, 
+        monoallelic: bool, 
+        AS_VQSLOD: float64, 
+        InbreedingCoeff: float64, 
+        vrs: struct {
+            VRS_Allele_IDs: array<str>, 
+            VRS_Starts: array<int32>, 
+            VRS_Ends: array<int32>, 
+            VRS_States: array<str>
+        }
+    } 
+    'vep': struct {
+        allele_string: str, 
+        end: int32, 
+        id: str, 
+        input: str, 
+        intergenic_consequences: array<struct {
+            allele_num: int32, 
+            consequence_terms: array<str>, 
+            impact: str, 
+            variant_allele: str
+        }>, 
+        most_severe_consequence: str, 
+        motif_feature_consequences: array<struct {
+            allele_num: int32, 
+            consequence_terms: array<str>, 
+            high_inf_pos: str, 
+            impact: str, 
+            motif_feature_id: str, 
+            motif_name: str, 
+            motif_pos: int32, 
+            motif_score_change: float64, 
+            transcription_factors: array<str>, 
+            strand: int32, 
+            variant_allele: str
+        }>, 
+        regulatory_feature_consequences: array<struct {
+            allele_num: int32, 
+            biotype: str, 
+            consequence_terms: array<str>, 
+            impact: str, 
+            regulatory_feature_id: str, 
+            variant_allele: str
+        }>, 
+        seq_region_name: str, 
+        start: int32, 
+        strand: int32, 
+        transcript_consequences: array<struct {
+            allele_num: int32, 
+            amino_acids: str, 
+            appris: str, 
+            biotype: str, 
+            canonical: int32, 
+            ccds: str, 
+            cdna_start: int32, 
+            cdna_end: int32, 
+            cds_end: int32, 
+            cds_start: int32, 
+            codons: str, 
+            consequence_terms: array<str>, 
+            distance: int32, 
+            domains: array<struct {
+                db: str, 
+                name: str
+            }>, 
+            exon: str, 
+            flags: str, 
+            gene_id: str, 
+            gene_pheno: int32, 
+            gene_symbol: str, 
+            gene_symbol_source: str, 
+            hgnc_id: str, 
+            hgvsc: str, 
+            hgvsp: str, 
+            hgvs_offset: int32, 
+            impact: str, 
+            intron: str, 
+            lof: str, 
+            lof_flags: str, 
+            lof_filter: str, 
+            lof_info: str, 
+            mane_select: str, 
+            mane_plus_clinical: str, 
+            mirna: array<str>, 
+            polyphen_prediction: str, 
+            polyphen_score: float64, 
+            protein_end: int32, 
+            protein_start: int32, 
+            protein_id: str, 
+            sift_prediction: str, 
+            sift_score: float64, 
+            source: str, 
+            strand: int32, 
+            transcript_id: str, 
+            tsl: int32, 
+            uniprot_isoform: array<str>, 
+            variant_allele: str
+        }>, 
+        variant_class: str
+    } 
+    'rf': struct {
+        rf_positive_label: bool, 
+        rf_negative_label: bool, 
+        rf_label: str, 
+        rf_train: bool, 
+        rf_tp_probability: float64
+    } 
+    'region_flag': struct {
+        lcr: bool, 
+        segdup: bool, 
+        non_par: bool
+    } 
+    'allele_info': struct {
+        variant_type: str, 
+        allele_type: str, 
+        n_alt_alleles: int32, 
+        was_mixed: bool
+    } 
+    'histograms': struct {
+        qual_hists: struct {
+            gq_hist_all: struct {
+                bin_edges: array<float64>, 
+                bin_freq: array<int64>, 
+                n_smaller: int64, 
+                n_larger: int64
+            }, 
+            dp_hist_all: struct {
+                bin_edges: array<float64>, 
+                bin_freq: array<int64>, 
+                n_smaller: int64, 
+                n_larger: int64
+            }, 
+            gq_hist_alt: struct {
+                bin_edges: array<float64>, 
+                bin_freq: array<int64>, 
+                n_smaller: int64, 
+                n_larger: int64
+            }, 
+            dp_hist_alt: struct {
+                bin_edges: array<float64>, 
+                bin_freq: array<int64>, 
+                n_smaller: int64, 
+                n_larger: int64
+            }, 
+            ab_hist_alt: struct {
+                bin_edges: array<float64>, 
+                bin_freq: array<int64>, 
+                n_smaller: int64, 
+                n_larger: int64
+            }
+        }, 
+        raw_qual_hists: struct {
+            gq_hist_all: struct {
+                bin_edges: array<float64>, 
+                bin_freq: array<int64>, 
+                n_smaller: int64, 
+                n_larger: int64
+            }, 
+            dp_hist_all: struct {
+                bin_edges: array<float64>, 
+                bin_freq: array<int64>, 
+                n_smaller: int64, 
+                n_larger: int64
+            }, 
+            gq_hist_alt: struct {
+                bin_edges: array<float64>, 
+                bin_freq: array<int64>, 
+                n_smaller: int64, 
+                n_larger: int64
+            }, 
+            dp_hist_alt: struct {
+                bin_edges: array<float64>, 
+                bin_freq: array<int64>, 
+                n_smaller: int64, 
+                n_larger: int64
+            }, 
+            ab_hist_alt: struct {
+                bin_edges: array<float64>, 
+                bin_freq: array<int64>, 
+                n_smaller: int64, 
+                n_larger: int64
+            }
+        }, 
+        age_hists: array<struct {
+            age_hist_hom: struct {
+                bin_edges: array<float64>, 
+                bin_freq: array<int64>, 
+                n_smaller: int64, 
+                n_larger: int64
+            }, 
+            age_hist_ht: struct {
+                bin_edges: array<float64>, 
+                bin_freq: array<int64>, 
+                n_smaller: int64, 
+                n_larger: int64
+            }
+        }>
+    } 
+    'in_silico_predictors': struct {
+        cadd: struct {
+            phred: float32, 
+            raw_score: float32, 
+            has_duplicate: bool
+        }, 
+        revel: struct {
+            revel_score: float64, 
+            has_duplicate: bool
+        }, 
+        splice_ai: struct {
+            splice_ai_score: float32, 
+            splice_consequence: str, 
+            has_duplicate: bool
+        }, 
+        pangolin: struct {
+            pangolin_score: float64
+        }
+    } 
+----------------------------------------
+Key: ['locus', 'alleles']
+----------------------------------------
\ No newline at end of file
diff --git a/data-pipeline/schemas/gnomad_v4_variants/prepare_gnomad_v4_exome_variants/output/gnomad_v4_exome_variants_base.ht.schema b/data-pipeline/schemas/gnomad_v4_variants/prepare_gnomad_v4_exome_variants/output/gnomad_v4_exome_variants_base.ht.schema
new file mode 100644
index 000000000..89fba5a59
--- /dev/null
+++ b/data-pipeline/schemas/gnomad_v4_variants/prepare_gnomad_v4_exome_variants/output/gnomad_v4_exome_variants_base.ht.schema
@@ -0,0 +1,334 @@
+----------------------------------------
+Global fields:
+    'freq_meta': array<dict<str, str>> 
+    'freq_index_dict': dict<str, int32> 
+    'faf_meta': array<dict<str, str>> 
+    'faf_index_dict': dict<str, int32> 
+    'freq_sample_count': array<int32> 
+    'filtering_model': struct {
+        model_name: str, 
+        score_name: str, 
+        feature_medians: dict<tuple (
+            str
+        ), struct {
+            a_index: int32, 
+            n_alt_alleles: int32, 
+            AS_pab_max: float64, 
+            AS_MQRankSum: float64, 
+            AS_SOR: float64, 
+            AS_ReadPosRankSum: float64
+        }>, 
+        variants_by_strata: dict<tuple (
+            str
+        ), int64>, 
+        features_importance: dict<str, float64>, 
+        features: array<str>, 
+        test_results: array<struct {
+            rf_prediction: str, 
+            rf_label: str, 
+            n: int32
+        }>, 
+        rf_snv_cutoff: struct {
+            bin: float64, 
+            min_score: float64
+        }, 
+        rf_indel_cutoff: struct {
+            bin: float64, 
+            min_score: float64
+        }, 
+        inbreeding_cutoff: float64, 
+        model_id: str
+    } 
+    'tool_versions': struct {
+        dbsnp_version: str, 
+        cadd_version: str, 
+        revel_version: str, 
+        splicaai_version: str, 
+        primateai_version: str, 
+        pangolin_version: str, 
+        vrs_version: str
+    } 
+    'vep_globals': struct {
+        vep_version: str, 
+        vep_csq_header: str, 
+        vep_help: str, 
+        vep_config: str
+    } 
+    'age_distribution': struct {
+        bin_edges: array<float64>, 
+        bin_freq: array<int32>, 
+        n_smaller: int32, 
+        n_larger: int32
+    } 
+    'age_index_dict': dict<str, int32> 
+    'age_meta': array<dict<str, str>> 
+    'grpmax_index_dict': dict<str, int32> 
+    'grpmax_meta': array<dict<str, str>> 
+    'README': dict<str, str> 
+    'gnomad_qc_repo': str 
+    'gnomad_methods_repo': str 
+----------------------------------------
+Row fields:
+    'locus': locus<GRCh38> 
+    'alleles': array<str> 
+    'grpmax': array<struct {
+        AC: int32, 
+        AF: float64, 
+        AN: int32, 
+        homozygote_count: int32, 
+        grp: str, 
+        faf95: float64
+    }> 
+    'rsids': set<str> 
+    'vep': struct {
+        allele_string: str, 
+        end: int32, 
+        id: str, 
+        input: str, 
+        intergenic_consequences: array<struct {
+            allele_num: int32, 
+            consequence_terms: array<str>, 
+            impact: str, 
+            variant_allele: str
+        }>, 
+        most_severe_consequence: str, 
+        motif_feature_consequences: array<struct {
+            allele_num: int32, 
+            consequence_terms: array<str>, 
+            high_inf_pos: str, 
+            impact: str, 
+            motif_feature_id: str, 
+            motif_name: str, 
+            motif_pos: int32, 
+            motif_score_change: float64, 
+            transcription_factors: array<str>, 
+            strand: int32, 
+            variant_allele: str
+        }>, 
+        regulatory_feature_consequences: array<struct {
+            allele_num: int32, 
+            biotype: str, 
+            consequence_terms: array<str>, 
+            impact: str, 
+            regulatory_feature_id: str, 
+            variant_allele: str
+        }>, 
+        seq_region_name: str, 
+        start: int32, 
+        strand: int32, 
+        transcript_consequences: array<struct {
+            allele_num: int32, 
+            amino_acids: str, 
+            appris: str, 
+            biotype: str, 
+            canonical: int32, 
+            ccds: str, 
+            cdna_start: int32, 
+            cdna_end: int32, 
+            cds_end: int32, 
+            cds_start: int32, 
+            codons: str, 
+            consequence_terms: array<str>, 
+            distance: int32, 
+            domains: array<struct {
+                db: str, 
+                name: str
+            }>, 
+            exon: str, 
+            flags: str, 
+            gene_id: str, 
+            gene_pheno: int32, 
+            gene_symbol: str, 
+            gene_symbol_source: str, 
+            hgnc_id: str, 
+            hgvsc: str, 
+            hgvsp: str, 
+            hgvs_offset: int32, 
+            impact: str, 
+            intron: str, 
+            lof: str, 
+            lof_flags: str, 
+            lof_filter: str, 
+            lof_info: str, 
+            mane_select: str, 
+            mane_plus_clinical: str, 
+            mirna: array<str>, 
+            polyphen_prediction: str, 
+            polyphen_score: float64, 
+            protein_end: int32, 
+            protein_start: int32, 
+            protein_id: str, 
+            sift_prediction: str, 
+            sift_score: float64, 
+            source: str, 
+            strand: int32, 
+            transcript_id: str, 
+            tsl: int32, 
+            uniprot_isoform: array<str>, 
+            variant_allele: str
+        }>, 
+        variant_class: str
+    } 
+    'rf': struct {
+        rf_positive_label: bool, 
+        rf_negative_label: bool, 
+        rf_label: str, 
+        rf_train: bool, 
+        rf_tp_probability: float64
+    } 
+    'in_silico_predictors': struct {
+        cadd: struct {
+            phred: float32, 
+            raw_score: float32, 
+            has_duplicate: bool
+        }, 
+        revel: struct {
+            revel_score: float64, 
+            has_duplicate: bool
+        }, 
+        splice_ai: struct {
+            splice_ai_score: float32, 
+            splice_consequence: str, 
+            has_duplicate: bool
+        }, 
+        pangolin: struct {
+            pangolin_score: float64
+        }
+    } 
+    'variant_id': str 
+    'colocated_variants': struct {
+        all: array<str>, 
+        non_ukb: array<str>
+    } 
+    'gnomad': struct {
+        freq: struct {
+            all: struct {
+                ac: int32, 
+                ac_raw: int32, 
+                an: int32, 
+                hemizygote_count: int32, 
+                homozygote_count: int32, 
+                populations: array<struct {
+                    id: str, 
+                    ac: int32, 
+                    an: int32, 
+                    hemizygote_count: int32, 
+                    homozygote_count: int32
+                }>
+            }, 
+            non_ukb: struct {
+                ac: int32, 
+                ac_raw: int32, 
+                an: int32, 
+                hemizygote_count: int32, 
+                homozygote_count: int32, 
+                populations: array<struct {
+                    id: str, 
+                    ac: int32, 
+                    an: int32, 
+                    hemizygote_count: int32, 
+                    homozygote_count: int32
+                }>
+            }
+        }, 
+        faf95: struct {
+            popmax: float64, 
+            popmax_population: str
+        }, 
+        faf99: struct {
+            popmax: float64, 
+            popmax_population: str
+        }, 
+        age_distribution: struct {
+            het: array<struct {
+                bin_edges: array<float64>, 
+                bin_freq: array<int64>, 
+                n_smaller: int64, 
+                n_larger: int64
+            }>, 
+            hom: array<struct {
+                bin_edges: array<float64>, 
+                bin_freq: array<int64>, 
+                n_smaller: int64, 
+                n_larger: int64
+            }>
+        }, 
+        filters: set<str>, 
+        quality_metrics: struct {
+            allele_balance: struct {
+                alt_adj: struct {
+                    bin_edges: array<float64>, 
+                    bin_freq: array<int64>, 
+                    n_smaller: int64, 
+                    n_larger: int64
+                }, 
+                alt_raw: struct {
+                    bin_edges: array<float64>, 
+                    bin_freq: array<int64>, 
+                    n_smaller: int64, 
+                    n_larger: int64
+                }
+            }, 
+            genotype_depth: struct {
+                all_adj: struct {
+                    bin_edges: array<float64>, 
+                    bin_freq: array<int64>, 
+                    n_smaller: int64, 
+                    n_larger: int64
+                }, 
+                all_raw: struct {
+                    bin_edges: array<float64>, 
+                    bin_freq: array<int64>, 
+                    n_smaller: int64, 
+                    n_larger: int64
+                }, 
+                alt_adj: struct {
+                    bin_edges: array<float64>, 
+                    bin_freq: array<int64>, 
+                    n_smaller: int64, 
+                    n_larger: int64
+                }, 
+                alt_raw: struct {
+                    bin_edges: array<float64>, 
+                    bin_freq: array<int64>, 
+                    n_smaller: int64, 
+                    n_larger: int64
+                }
+            }, 
+            genotype_quality: struct {
+                all_adj: struct {
+                    bin_edges: array<float64>, 
+                    bin_freq: array<int64>, 
+                    n_smaller: int64, 
+                    n_larger: int64
+                }, 
+                all_raw: struct {
+                    bin_edges: array<float64>, 
+                    bin_freq: array<int64>, 
+                    n_smaller: int64, 
+                    n_larger: int64
+                }, 
+                alt_adj: struct {
+                    bin_edges: array<float64>, 
+                    bin_freq: array<int64>, 
+                    n_smaller: int64, 
+                    n_larger: int64
+                }, 
+                alt_raw: struct {
+                    bin_edges: array<float64>, 
+                    bin_freq: array<int64>, 
+                    n_smaller: int64, 
+                    n_larger: int64
+                }
+            }, 
+            site_quality_metrics: array<struct {
+                metric: str, 
+                value: float64
+            }>
+        }
+    } 
+    'subsets': set<str> 
+    'flags': set<str> 
+----------------------------------------
+Key: ['locus', 'alleles']
+----------------------------------------
\ No newline at end of file
diff --git a/data-pipeline/src/data_pipeline/datasets/gnomad_v4/gnomad_v4_variants.py b/data-pipeline/src/data_pipeline/datasets/gnomad_v4/gnomad_v4_variants.py
index cbd421034..2b8314b8e 100644
--- a/data-pipeline/src/data_pipeline/datasets/gnomad_v4/gnomad_v4_variants.py
+++ b/data-pipeline/src/data_pipeline/datasets/gnomad_v4/gnomad_v4_variants.py
@@ -15,8 +15,8 @@ def freq_index_key(subset=None, pop=None, sex=None, raw=False):
     return "-".join(parts)
 
 
-def prepare_gnomad_v4_variants(path):
-    ds = hl.read_table(path)
+def prepare_gnomad_v4_variants(input_path):
+    ds = hl.read_table(input_path)
     g = hl.eval(ds.globals)
 
     subsets = set(m.get("subset", None) for m in g.freq_meta)
@@ -84,7 +84,7 @@ def subset_filter(subset):
     ds = ds.annotate(in_autosome_or_par=ds.locus.in_autosome_or_par())
 
     ds = ds.annotate(
-        exome=hl.struct(
+        gnomad=hl.struct(
             freq=hl.struct(
                 **{
                     subset
@@ -124,15 +124,15 @@ def subset_filter(subset):
 
     # If a variant is not present in a subset, do not store population frequencies for that subset
     ds = ds.annotate(
-        exome=ds.exome.annotate(
-            freq=ds.exome.freq.annotate(
+        gnomad=ds.gnomad.annotate(
+            freq=ds.gnomad.freq.annotate(
                 **{
                     subset
-                    or "all": ds.exome.freq[subset or "all"].annotate(
+                    or "all": ds.gnomad.freq[subset or "all"].annotate(
                         populations=hl.if_else(
-                            ds.exome.freq[subset or "all"].ac_raw == 0,
-                            hl.empty_array(ds.exome.freq[subset or "all"].populations.dtype.element_type),
-                            ds.exome.freq[subset or "all"].populations,
+                            ds.gnomad.freq[subset or "all"].ac_raw == 0,
+                            hl.empty_array(ds.gnomad.freq[subset or "all"].populations.dtype.element_type),
+                            ds.gnomad.freq[subset or "all"].populations,
                         )
                     )
                     for subset in subsets
@@ -149,7 +149,7 @@ def subset_filter(subset):
 
     ds = ds.annotate(
         subsets=hl.set(
-            hl.array([(subset, ds.exome.freq[subset].ac_raw > 0) for subset in subsets if subset is not None])
+            hl.array([(subset, ds.gnomad.freq[subset].ac_raw > 0) for subset in subsets if subset is not None])
             .filter(lambda t: t[1])
             .map(lambda t: t[0])
         )
@@ -163,7 +163,7 @@ def subset_filter(subset):
 
     # Get popmax FAFs
     ds = ds.annotate(
-        exome=ds.exome.annotate(
+        gnomad=ds.gnomad.annotate(
             faf95=hl.rbind(
                 hl.sorted(
                     hl.array(
@@ -196,7 +196,7 @@ def subset_filter(subset):
                     hl.struct(popmax=hl.null(hl.tfloat), popmax_population=hl.null(hl.tstr)),
                 ),
             ),
-        )
+        ),
     )
 
     ds = ds.drop("faf")
@@ -206,7 +206,7 @@ def subset_filter(subset):
     ####################
 
     ds = ds.annotate(
-        exome=ds.exome.annotate(
+        gnomad=ds.gnomad.annotate(
             age_distribution=hl.struct(
                 het=ds.histograms.age_hists.age_hist_ht, hom=ds.histograms.age_hists.age_hist_hom
             )
@@ -218,7 +218,7 @@ def subset_filter(subset):
     ###################
 
     ds = ds.annotate(
-        exome=ds.exome.annotate(
+        gnomad=ds.gnomad.annotate(
             filters=ds.filters,
             quality_metrics=hl.struct(
                 allele_balance=hl.struct(
diff --git a/data-pipeline/src/data_pipeline/datasets/gnomad_v4/types/initial_variant.py b/data-pipeline/src/data_pipeline/datasets/gnomad_v4/types/initial_variant.py
index 0e4e36978..2d810ef5a 100644
--- a/data-pipeline/src/data_pipeline/datasets/gnomad_v4/types/initial_variant.py
+++ b/data-pipeline/src/data_pipeline/datasets/gnomad_v4/types/initial_variant.py
@@ -130,69 +130,69 @@ class Domain:
 
 @attr.define
 class TranscriptConsequence:
-    allele_num: int
-    amino_acids: str
-    appris: str
-    biotype: str
+    allele_num: Union[int, None]
+    amino_acids: Union[str, None]
+    appris: Union[str, None]
+    biotype: Union[str, None]
     canonical: Union[int, None]
-    ccds: str
+    ccds: Union[str, None]
     cdna_start: Union[int, None]
     cdna_end: Union[int, None]
     cds_end: Union[int, None]
     cds_start: Union[int, None]
-    codons: str
+    codons: Union[str, None]
     consequence_terms: List[str]
-    distance: int
+    distance: Union[int, None]
     domains: Union[List[Domain], None]
-    exon: str
-    flags: str
-    gene_id: str
+    exon: Union[str, None]
+    flags: Union[str, None]
+    gene_id: Union[str, None]
     gene_pheno: Union[int, None]
-    gene_symbol: str
-    gene_symbol_source: str
-    hgnc_id: str
-    hgvsc: str
-    hgvsp: str
+    gene_symbol: Union[str, None]
+    gene_symbol_source: Union[str, None]
+    hgnc_id: Union[str, None]
+    hgvsc: Union[str, None]
+    hgvsp: Union[str, None]
     hgvs_offset: Union[int, None]
-    impact: str
-    intron: str
-    lof: str
-    lof_flags: str
-    lof_filter: str
-    lof_info: str
-    mane_select: str
-    mane_plus_clinical: str
+    impact: Union[str, None]
+    intron: Union[str, None]
+    lof: Union[str, None]
+    lof_flags: Union[str, None]
+    lof_filter: Union[str, None]
+    lof_info: Union[str, None]
+    mane_select: Union[str, None]
+    mane_plus_clinical: Union[str, None]
     mirna: Union[List[str], None]
-    polyphen_prediction: str
+    polyphen_prediction: Union[str, None]
     polyphen_score: Union[float, None]
     protein_end: Union[int, None]
     protein_start: Union[int, None]
-    protein_id: str
+    protein_id: Union[str, None]
     sift_prediction: Union[str, None]
     sift_score: Union[float, None]
-    source: str
-    strand: int
-    transcript_id: str
+    source: Union[str, None]
+    strand: Union[int, None]
+    transcript_id: Union[str, None]
     tsl: Union[int, None]
     uniprot_isoform: Union[List[str], None]
-    variant_allele: str
+    variant_allele: Union[str, None]
 
 
 @attr.define
 class Vep:
-    allele_string: str
-    end: int
-    id: str
-    input: str
+    allele_string: Union[str, None]
+    end: Union[int, None]
+    id: Union[str, None]
+    input: Union[str, None]
     intergenic_consequences: Union[List[Consequence], None]
-    most_severe_consequence: str
+    most_severe_consequence: Union[str, None]
     motif_feature_consequences: Union[List[Consequence], None]
     regulatory_feature_consequences: Union[List[Consequence], None]
-    seq_region_name: str
-    start: int
-    strand: int
+    seq_region_name: Union[str, None]
+    start: Union[int, None]
+    strand: Union[int, None]
     transcript_consequences: List[TranscriptConsequence]
-    variant_class: str
+    variant_class: Union[str, None]
 
 
 @attr.define
diff --git a/data-pipeline/src/data_pipeline/pipelines/gnomad_v4_variants.py b/data-pipeline/src/data_pipeline/pipelines/gnomad_v4_variants.py
index 0e53ea055..6da147d37 100644
--- a/data-pipeline/src/data_pipeline/pipelines/gnomad_v4_variants.py
+++ b/data-pipeline/src/data_pipeline/pipelines/gnomad_v4_variants.py
@@ -1,51 +1,64 @@
 from data_pipeline.pipeline import Pipeline, run_pipeline
 
-from data_pipeline.data_types.variant import annotate_variants, annotate_transcript_consequences
-
+from data_pipeline.config import config
 from data_pipeline.datasets.gnomad_v4.gnomad_v4_variants import prepare_gnomad_v4_variants
 
-from data_pipeline.pipelines.gnomad_v4_coverage import pipeline as coverage_pipeline
-from data_pipeline.pipelines.genes import pipeline as genes_pipeline
+# from data_pipeline.data_types.variant import annotate_variants, annotate_transcript_consequences
 
+# from data_pipeline.pipelines.gnomad_v4_coverage import pipeline as coverage_pipeline
+# from data_pipeline.pipelines.genes import pipeline as genes_pipeline
 
-pipeline = Pipeline()
 
-pipeline.add_task(
-    "prepare_gnomad_v4_variants",
-    prepare_gnomad_v4_variants,
-    "/gnomad_v4/gnomad_v4_variants_base.ht",
-    {"path": "gs://gnomad-matt-data-pipeline/external_sources/2023-09-07-exome-variants-v4-mock/mock_v4_release.ht"},
-)
+pipeline = Pipeline(name="gnomad_v4_variants")
 
 pipeline.add_task(
-    "annotate_gnomad_v4_variants",
-    annotate_variants,
-    "/gnomad_v4/gnomad_v4_variants_annotated_1.ht",
-    {
-        "variants_path": pipeline.get_task("prepare_gnomad_v4_variants"),
-        # We need to subset regions chr1:10030:10150
-        "exome_coverage_path": coverage_pipeline.get_output("exome_coverage"),
-        "genome_coverage_path": coverage_pipeline.get_output("genome_coverage"),
-        # "caids_path": "gs://gnomad-browser-data-pipeline/caids/gnomad_v4_caids.ht",
+    name="prepare_gnomad_v4_exome_variants",
+    task_function=prepare_gnomad_v4_variants,
+    output_path="/gnomad_v4/gnomad_v4_exome_variants_base.ht",
+    inputs={
+        "input_path": "external_datasets/mock_v4_release.ht",
     },
+    # params={"sequencing_type": "exome"},
 )
 
-pipeline.add_task(
-    "annotate_gnomad_v4_transcript_consequences",
-    annotate_transcript_consequences,
-    "/gnomad_v4/gnomad_v4_variants_annotated_2.ht",
-    {
-        "variants_path": pipeline.get_task("annotate_gnomad_v4_variants"),
-        "transcripts_path": genes_pipeline.get_output("base_transcripts_grch38"),
-        "mane_transcripts_path": genes_pipeline.get_output("mane_select_transcripts"),
-    },
-)
+# pipeline.add_task(
+#     "prepare_gnomad_v4_genome_variants",
+#     prepare_gnomad_v4_variants,
+#     "/gnomad_v4/gnomad_v4_genome_variants_base.ht",
+#     {
+#         "path": "gs://gnomad-matt-data-pipeline/external_sources/2023-09-07-exome-variants-v4-mock/mock_v4_release.ht",
+#         "type": "genome",
+#     },
+# )
+
+# pipeline.add_task(
+#     "annotate_gnomad_v4_variants",
+#     annotate_variants,
+#     "/gnomad_v4/gnomad_v4_variants_annotated_1.ht",
+#     {
+#         "variants_path": pipeline.get_task("prepare_gnomad_v4_exome_variants"),
+#         "exome_coverage_path": coverage_pipeline.get_output("exome_coverage"),
+#         "genome_coverage_path": coverage_pipeline.get_output("genome_coverage"),
+#         # "caids_path": "gs://gnomad-browser-data-pipeline/caids/gnomad_v4_caids.ht",
+#     },
+# )
+
+# pipeline.add_task(
+#     "annotate_gnomad_v4_transcript_consequences",
+#     annotate_transcript_consequences,
+#     "/gnomad_v4/gnomad_v4_variants_annotated_2.ht",
+#     {
+#         "variants_path": pipeline.get_task("annotate_gnomad_v4_variants"),
+#         "transcripts_path": genes_pipeline.get_output("base_transcripts_grch38"),
+#         "mane_transcripts_path": genes_pipeline.get_output("mane_select_transcripts"),
+#     },
+# )
 
 ###############################################
 # Outputs
 ###############################################
 
-pipeline.set_outputs({"variants": "annotate_gnomad_v4_transcript_consequences"})
+# pipeline.set_outputs({"variants": "annotate_gnomad_v4_transcript_consequences"})
 
 ###############################################
 # Run