Skip to content

Commit c126ca9

Browse files
Create README.md
1 parent 6d8c9ff commit c126ca9

1 file changed

Lines changed: 35 additions & 0 deletions

File tree

latent_embeddings/README.md

Lines changed: 35 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,35 @@
1+
### Bichrom Embeddings
2+
This module will report latent (2-D) co-ordinates for any DNA sequence and chromatin feature combination, given a trained Bichrom model.
3+
4+
**(see trainNN README.md for details on how to train a Bichrom model)**
5+
6+
### Usage
7+
8+
To view help: ```python embed.py --help```
9+
```
10+
usage: embed.py [-h] -model MODEL -seq SEQ -chrom CHROM [CHROM ...] -length
11+
LENGTH -outdir OUTDIR
12+
13+
Derives and plots 2-D embeddings
14+
15+
optional arguments:
16+
-h, --help show this help message and exit
17+
-model MODEL Trained bichrom model
18+
-seq SEQ Sequence file, each line contains sequences associated
19+
with one genomic window
20+
-chrom CHROM [CHROM ...]
21+
List of files, each containing chromatin features,
22+
associated with one genomic window per line
23+
-length LENGTH Length of training windows
24+
-outdir OUTDIR Output directory
25+
```
26+
27+
### Input Data Requirements and Details
28+
29+
* The **sequence file** contains: Sequences of length L. (Permitted bases: A, T, G, C, N)
30+
* The **chromatin files** contains: A binned vector of chromatin signal.
31+
32+
(Example: If **L=500** and binsize **B=10**, then chromatin signal vector must contain **L/B** values)
33+
34+
Note: Multiple chromatin track files can be provided with the -chrom argument (For example: ATAC-seq, H3K27ac, ... etc.).
35+

0 commit comments

Comments
 (0)