Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
37 changes: 5 additions & 32 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -20,37 +20,10 @@ infrastructure.

# coderdata Data Model

The goal of the data model is to collate drug response data together with molecular data in a way that can be easily ingested by machine learning models. The overall schema is shown below.
The coderdata schema is maintained in [LinkML](schema/coderdata.yaml)
and can be udpated via a commit to the repository. For more details,
please see the [schema description](schema/README.md).

<img src="origDataSchema.jpg" width=25% height=25%>
## Building the data package

We will store the data in tables that are represented by the files below. Each data-specific model can be generated from a smaller set of these tables. The schema for these tables is represented below.

For each dataset added, the files are comma-delimited and named follows:
1. genes.csv
2. drugs.tsv.gz --> Drug names have commas and quotes in them, therefore require tab delimited
3. samples.csv
4. experiments.csv.gz --> compressed to fit on github
5. transcriptomics.csv.gz
6. mutations.csv.gz
7. copy_number.csv.gz
8. methylation.csv.gz
9. mirnas.csv.gz

## Building the data model

Below is a description of how the data model is built.

| Data model step | Description/Dependencies | Script | Destination |
| --- | --- | --- | --- |
| Build cell line data | Runs through PGX and existing CCLE data to compile all values | [cell_line/buildInitialDataset.py](cell_line/buildInitialDataset.py) | [./cell_line] |
| Build cptac data | This uses the genes files created in the [./cell_line] directory but generates additional samples. | [cptac/getCptacData.py](cptac/getCptacData.py) | [./cptac] |
| Get HCMI data | This uses a fixed manifest to download the data into the proper schema | [hcmi/getHCMIData.py](hcmi/getHCMIData.py) | [./hcmi] |


## Current data
What data is stored here?

## Using the data model

Files are stored on FigShare. We need to build a script that pulls those data as needed.
The data package is currently assembled via continuous automation,
9 changes: 9 additions & 0 deletions schema/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -18,9 +18,18 @@ model. Currently we have the following objects in our model:
| Gene | Human gene identifiers|
| Samples | Samples in which omics and drugs are measured |
| Drugs | Drugs evaluated on samples |
| Drug Descriptor | Computational descriptions of drug properties |
| Transcriptomics | Gene exrpression data |
| Mutations | Gene mutation data |
| Copy Number | Copy number data |
| Proteomics | Proteomics data|
| Experiments | Drug experiments that measure the viability of the cell |
| Perturbations | Experiments that measure molecular changes of the cell |
| Combinations | Synergy calculations from pairs of drugs


### Schema image

The data model is shown here:
<img src="coderdataSchema.jpg" width=25% height=25%>

86 changes: 57 additions & 29 deletions schema/coderdata.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -91,6 +91,17 @@ classes:
range: float
pubchem_id:
description: PubChem Identifier for this drug, can be many.
Drug Descriptor:
description: Computational summary of drug chemical properties
slots:
- improve_drug_id
attributes:
structural_descriptor:
description: string name describing structural descriptor
descriptor_value:
range: linkml: Any
description: value representing descriptor value

Transcriptomics:
description:
slots:
Expand Down Expand Up @@ -146,39 +157,19 @@ classes:
- time
- time_unit
attributes:
auc:
range: float
description: Area under the fitted curve
ic50:
range: float
description: Concetration at which inhibitor binding is reduced by half
ec50:
range: float
description: Concentration required to have half cells with adverse effect
ec50se:
range: float
description: Standard error in EC50 calculation
einf:
range: float
description: Fraction of cells not susceptible to the drug
hs:
range: float
description: Hill slop binding cooperativity
aac1:
range: float
description: Area above curve measured by integrating the data points
auc1:
range: float
description: Area under the curve measured by integrating data points
dss1:
range: float
description: Check on this
dose_response_metric:
range: CurveMetric
description: Metric by which dose response value is measured
dose_response_value:
description: Value of metric
Perturbations:
slots:
- entrez_id
- improve_sample_id
- time
- time_unit
- source
- study
attributes:
data_value:
range: float
Expand All @@ -191,8 +182,43 @@ classes:
perturbation_type:
description: The perturbagen that was used
range: CellPerturbation

Combinations:
slots:
- improve_sample_id
- source
- study
attributes:
drug_one_id:
description: improve_drug_id of first drug
drug_two_id:
description: imrrove_drug_id of second drug
synergy_metric:
description: metric calculated
synergy_value:
description: value of metric
enums:
ResponseMetric:
permissible_values:
fit auc:
description: Area under the fitted curve
fit ic50:
description: Concetration at which inhibitor binding is reduced by half
fit ec50:
description: Concentration required to have half cells with adverse effect
fit ec50se:
description: Standard error in EC50 calculation
fit einf:
description: Fraction of cells not susceptible to the drug
fit hs:
description: Hill slop binding cooperativity
fit r2:
description: R2 value between fit and actual data
aac:
description: Area above curve measured by integrating the data points
auc:
description: Area under the curve measured by integrating data points
dss:
description: I believe this is the drug sensitivity score
CellPerturbation:
permissible_values:
gene_ko:
Expand All @@ -207,6 +233,7 @@ enums:
Variant:
permissible_values:
3'UTR:
3'Flank:
5'Flank:
5'UTR:
Undetermined:
Expand All @@ -228,8 +255,9 @@ enums:
Stop_Codon_Del:
Stop_Codon_Ins:
Silent:
Splice_Region:
Splice_Site:
Translationn_Start_Site:
Translation_Start_Site:
CopyCall:
permissible_values:
amp:
Expand Down
Binary file added schema/coderdataSchema.jpg
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.