expression/SOURCES.md: the prose still said two sources
Browse filesAdding hpa_consensus left the surrounding text stale -- "two references", "both sources
are measured RNA-seq", a key_type paragraph naming only GTEx and TCGA, and a ~105 MB
download size for a file that is now larger. Found by auditing every count claimed in the
published docs against the files themselves; the counts were right, the prose around them
was not.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
- expression/SOURCES.md +8 -7
expression/SOURCES.md
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
# `expression/` — sources
|
| 2 |
|
| 3 |
-
One file,
|
| 4 |
-
gene-keyed table and a peptide-keyed one are not the same object.
|
| 5 |
|
| 6 |
| | |
|
| 7 |
|---|---|
|
| 8 |
| file | `reference_expression.tsv.gz` |
|
| 9 |
| rows | 6,681,814 |
|
| 10 |
| schema | `key`, `key_type`, `source`, `context`, `median_tpm`, `q25_tpm`, `q75_tpm`, `n` |
|
| 11 |
-
| provenance | **experimental** (
|
| 12 |
|
| 13 |
| `source` | `key_type` | rows | contexts | what a row is |
|
| 14 |
|---|---|--:|--:|---|
|
|
@@ -18,12 +18,13 @@ gene-keyed table and a peptide-keyed one are not the same object.
|
|
| 18 |
|
| 19 |
Total 6,681,814 rows.
|
| 20 |
|
| 21 |
-
`key_type` is the join key and it differs by source on purpose: GTEx
|
| 22 |
-
normal tissue", TCGA answers "is this neoantigen transcribed in this tumour". A caller that treats
|
| 23 |
-
them as one column is asking a question
|
|
|
|
| 24 |
|
| 25 |
Read through `mhcmatch.expression` (`lookup`, `tissues`, `tumor_types`, `safety_profile`), which
|
| 26 |
-
keeps
|
| 27 |
|
| 28 |
---
|
| 29 |
|
|
|
|
| 1 |
# `expression/` — sources
|
| 2 |
|
| 3 |
+
One file, three references that are **never merged** — different measurements, different units, and
|
| 4 |
+
a gene-keyed table and a peptide-keyed one are not the same object.
|
| 5 |
|
| 6 |
| | |
|
| 7 |
|---|---|
|
| 8 |
| file | `reference_expression.tsv.gz` |
|
| 9 |
| rows | 6,681,814 |
|
| 10 |
| schema | `key`, `key_type`, `source`, `context`, `median_tpm`, `q25_tpm`, `q75_tpm`, `n` |
|
| 11 |
+
| provenance | **experimental** (all three sources are measured RNA-seq); the per-context median/IQR/`n` are **computed** summaries of it |
|
| 12 |
|
| 13 |
| `source` | `key_type` | rows | contexts | what a row is |
|
| 14 |
|---|---|--:|--:|---|
|
|
|
|
| 18 |
|
| 19 |
Total 6,681,814 rows.
|
| 20 |
|
| 21 |
+
`key_type` is the join key and it differs by source on purpose: GTEx and HPA answer "is this gene on
|
| 22 |
+
in normal tissue", TCGA answers "is this neoantigen transcribed in this tumour". A caller that treats
|
| 23 |
+
them as one column is asking a question none of them answers. GTEx and HPA are both gene-keyed but
|
| 24 |
+
are still not interchangeable — see the units caveat in the HPA section below.
|
| 25 |
|
| 26 |
Read through `mhcmatch.expression` (`lookup`, `tissues`, `tumor_types`, `safety_profile`), which
|
| 27 |
+
keeps them apart. Fetched by `mhcmatch bootstrap --reference` (~115 MB).
|
| 28 |
|
| 29 |
---
|
| 30 |
|