File size: 987 Bytes
0c7422e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
dd96f98
 
0c7422e
 
 
dd96f98
0c7422e
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
---
license: lgpl-3.0
language:
- en
tags:
- chemistry
- biology
---

# NucleoFind

Nucleic acid electron density interpretation remains a difficult problem for computer programs to deal with. Programs tend to rely on exhaustive searches to recognise characteristic features. NucleoFind is a deep-learning-based approach to interpreting and segmenting electron density. Using a crystallographic map, the positions of the phosphate group, sugar ring and nitrogenous base group are able to be predicted with high accuracy.
## Model Details

### Model Description

NucleoFind is based on a 3D-UNet architecture.


- **Developed by:** Jordan Dialpuri, Jon Agirre, Kathryn Cowtan and Paul Bond, York Structural Biology Laboratory, University of York
- **Funded by BBSRC and The Royal Society** 
- **Model type:** Multiclass
- **Language(s) (NLP):** Python
- **License:** LGPL-3

## Model Card Authors 

Jordan Dialpuri

## Model Card Contact

Jordan Dialpuri - jordan.dialpuri (at) york.ac.uk