Update README.md
Browse files
README.md
CHANGED
@@ -11,6 +11,8 @@ SegmentBorzoi is a segmentation model leveraging [Borzoi](https://www.biorxiv.or
|
|
11 |
elements in a sequence at a single nucleotide resolution. It was trained on 14 different classes, including gene (protein-coding genes, lncRNAs, 5’UTR, 3’UTR, exon, intron, splice acceptor and donor sites) and regulatory (polyA signal, tissue-invariant and
|
12 |
tissue-specific promoters and enhancers, and CTCF-bound sites) elements.
|
13 |
|
|
|
|
|
14 |
|
15 |
**Developed by:** [InstaDeep](https://huggingface.co/InstaDeepAI)
|
16 |
|
|
|
11 |
elements in a sequence at a single nucleotide resolution. It was trained on 14 different classes, including gene (protein-coding genes, lncRNAs, 5’UTR, 3’UTR, exon, intron, splice acceptor and donor sites) and regulatory (polyA signal, tissue-invariant and
|
12 |
tissue-specific promoters and enhancers, and CTCF-bound sites) elements.
|
13 |
|
14 |
+
We note that this model is based on the published implementation of Borzoi (originally in Keras). We used the [FlashZoi](https://github.com/johahi/borzoi-pytorch) implementation for the common parts to the published model for easier implementation,
|
15 |
+
but changed it accordingly to the published version when needed - such as the positional embeddings in the attention layers.
|
16 |
|
17 |
**Developed by:** [InstaDeep](https://huggingface.co/InstaDeepAI)
|
18 |
|