Location: /home/mohammadi/datasets/ (~1.5TB total)
Last audited: 2026-05-22
``
/home/mohammadi/datasets/
├── 01-ontologies/ ~75GB OWL/OBO files + registry.yaml
│ ├── owl/ Flat deduplicated OWL files (37+ ontologies)
│ ├── mappings/ SSSOM mapping files
│ ├── archive/ Legacy directory structure (pre-cleanup)
│ └── registry.yaml Single source of truth for all semantic resources
│
├── 02-vocabularies/ ~76GB UMLS 2026AA, SNOMED CT INT+US, MeSH
│
├── 03-knowledge-graphs/ ~440GB Monarch, PrimeKG, PlaNet, Open Targets, PKG2.0
│
├── 04-identifiers/ ~178GB Ensembl 115, UniChem, OLS4 SSSOM
│ └── ontologies/ SO, GO, HP, MP, etc. (OBO format)
│ └── so.obo ← Sequence Ontology (used by cytos)
│ └── seqrepo/ ← NOT YET INSTALLED (~50GB needed for VRS)
│
├── 05-annotations/ ~12GB CellxGene, HRA CCF, topic areas
│
├── 06-genotype/ ~103GB ALL genotype/WGS data
│ ├── personal/
│ │ ├── Shahin/
│ │ │ ├── NG1CNUL31H.mm2.sortdup.bqsr.hc.vcf.gz ← HC variant calls
│ │ │ ├── NG1CNUL31H.mm2.sortdup.bqsr.hc.vcf.gz.tbi
│ │ │ ├── NG1CNUL31H.mm2.sortdup.bqsr.cram ← Full read data
│ │ │ ├── NG1CNUL31H.mm2.sortdup.bqsr.cram.crai
│ │ │ ├── genotypes_ext.tsv ← Extended variant table
│ │ │ └── insertion_TBX1.fa ← TBX1 SV sequence
│ │ └── Olivia/
│ │ ├── *snp-indel.genome.vcf.gz ← SNP+INDEL
│ │ ├── *cnv.vcf.gz ← Copy number
│ │ └── *sv.vcf.gz ← Structural variants
│ ├── cohort/
│ │ └── PEC/
│ │ └── brainSCOPE_PEC_sample_genotypes_no_rna.vcf.gz ← 218 WGS samples
│ ├── gwas/
│ │ └── pgc/ 14 PGC psychiatric GWAS files (ADHD, ALCH, AN, ...)
│ ├── ldgm/
│ │ ├── index.duckdb ← DuckDB block index (1,360 entries)
│ │ ├── ancestry/EUR/ ← Raw edgelist + snplist files (graphLD native)
│ │ └── tiledb/EUR/ ← TileDB sparse Ω arrays (one per block)
│ └── reference/
│ └── gencode.v47.annotation.gtf.gz ← GENCODE v47 gene annotation
│
├── 07-single-cell/ TileDB-SOMA (PEC RNA-seq pending)
│ └── PEC/
│ ├── CMC/
│ ├── SZBDMulti-Seq/
│ ├── DevBrain-snRNAseq/
│ └── UCLA-ASD/
│
├── 08-neuroimaging/ ~319GB BIDS/NWB datasets
│ └── PEC/
│ └── synapse/genotype/WGS-Derived-ImputedGenotypes/ ← PEC RNA imputed VCFs
│
├── 09-literature/ BibTeX + PDF cache
├── 10-embeddings/ ~34GB Pre-computed embeddings
├── 11-benchmarks/ ~23GB Benchmark datasets
├── 12-network/ ~200GB Biological networks
├── 13-cell-lines/ Cell line data
├── 14-cohort-metadata/ Sample manifests + IRB + phenotypes
│ └── PEC/
│ └── SYNAPSE_METADATA_MANIFEST.tsv ← PEC sample manifest
└── 15-databases/ Database storage
└── surrealdb/
└── cytos.db ← SurrealDB persistent store
Location: /home/mohammadi/repos/cytognosis/cytos/tiledb/vcf/
(NOT in git — local disk only)
| Store | Samples | Source | Notes |
|-------|---------|--------|-------|
| Shahin/snp/ | 1 | HC VCF | 47 attrs including GT, DP, GQ, AD |Olivia/snp/
| | 1 | SNP+INDEL VCF | |Olivia/cnv/
| | 1 | CNV VCF | |Olivia/sv/
| | 1 | SV VCF | |PEC/snp/
| | 218 | brainSCOPE WGS | Full cohort |PEC_RNA/snp/
| | 226 | RNA-derived imputed | From 08-neuroimaging/PEC |
Location: /home/mohammadi/repos/cytognosis/cytos/data/gwas/
gwas_catalog/
├── SCZ_GCST90038608.h.tsv.gz 9.7M variants (SCZ, PMID 35396580)
├── BIP_GCST003724.h.tsv.gz 9.5M variants (BIP, PMID 21926972)
└── MDD_GCST006085.h.tsv.gz 20.0M variants (MDD, PMID 29892016)
`
PGC (minimal format, downloaded)
`
pgc/
├── ADHD.tsv.gz ├── AN.tsv.gz ├── ASD.tsv.gz
├── ALCH.tsv.gz ├── ANX.tsv.gz ├── BIP.tsv.gz
├── CANNABIS.tsv.gz ├── INSOMNIA.tsv.gz └── [more]
`
Known Data Issues
1. PEC WGS count: Expected 388 samples (paper), ingested 218. Remaining ~170 may be
from a different batch or protected-access tier. RNA-derived VCFs give 226 samples.
Not a data processing error — the source VCF has 218 samples.
2. seqrepo not installed: VRS ID computation requires seqrepo (~50GB).
Must install before Variant nodes can be created.
3. PEC RNA location: PEC RNA-derived VCFs are at
08-neuroimaging/PEC/ (not 07-single-cell/).
This is a pre-existing mis-organization; symlinks exist at both locations.
4. DVC tracked files: Some data is tracked by DVC with remote at GCS.
dvc status shows topic_areas and monarch_merge` stages have changed deps.