Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,86 @@
|
|
1 |
-
---
|
2 |
-
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
license:
|
5 |
+
- cc-by-4.0
|
6 |
+
size_categories:
|
7 |
+
- 1M<n<10M
|
8 |
+
tags:
|
9 |
+
- DNA Sequences
|
10 |
+
- Protein Sequences
|
11 |
+
- Computational Biology
|
12 |
+
- Bioinformatics
|
13 |
+
- Synthetic Biology
|
14 |
+
task_categories:
|
15 |
+
- token-classification
|
16 |
+
---
|
17 |
+
|
18 |
+
# CodonTransformer Dataset
|
19 |
+
|
20 |
+
A comprehensive collection of DNA and protein sequence pairs from a diverse range of organisms, suitable for various computational biology and bioinformatics applications.
|
21 |
+
|
22 |
+
## Dataset Summary
|
23 |
+
|
24 |
+
The CodonTransformer dataset is a large-scale compilation of 1,001,197 DNA sequences and their corresponding protein translations, sourced from 164 different organisms across Eukaryotes, Bacteria, and Archaea. This dataset provides a rich resource for studying gene sequences, codon usage, and protein expression across diverse species.
|
25 |
+
|
26 |
+
## Dataset Contents
|
27 |
+
|
28 |
+
- 1,001,197 DNA-protein sequence pairs
|
29 |
+
- Sequences from 164 organisms, including:
|
30 |
+
- Eukaryotes: Homo sapiens, Arabidopsis thaliana, Caenorhabditis elegans, Danio rerio, Drosophila melanogaster, Mus musculus, Saccharomyces cerevisiae
|
31 |
+
- Bacteria: Various Enterobacteriaceae species including Escherichia coli
|
32 |
+
- Archaea: Thermococcus barophilus, Sulfolobus solfataricus
|
33 |
+
- Chloroplast genomes: Chlamydomonas reinhardtii, Nicotiana tabacum
|
34 |
+
|
35 |
+
## Data Collection and Preprocessing
|
36 |
+
|
37 |
+
- **Source**: NCBI resources
|
38 |
+
- **Original Format**: Gene or CDS (Coding Sequence)
|
39 |
+
- **Protein Sequences**: Translated using NCBI Codon Tables
|
40 |
+
- **Quality Control**:
|
41 |
+
- DNA sequences divisible by three in length
|
42 |
+
- Start with a start codon
|
43 |
+
- End with a single stop codon
|
44 |
+
|
45 |
+
## Dataset Structure
|
46 |
+
|
47 |
+
Each entry contains:
|
48 |
+
- DNA sequence
|
49 |
+
- Corresponding protein sequence
|
50 |
+
- Organism information
|
51 |
+
|
52 |
+
## Uses and Applications
|
53 |
+
|
54 |
+
This dataset is valuable for various research areas and applications, including:
|
55 |
+
- Comparative genomics
|
56 |
+
- Codon usage analysis
|
57 |
+
- Protein expression optimization
|
58 |
+
- Synthetic biology and genetic engineering
|
59 |
+
- Machine learning models in bioinformatics
|
60 |
+
|
61 |
+
It has been used to train the CodonTransformer model for codon optimization tasks.
|
62 |
+
|
63 |
+
## Authors
|
64 |
+
Adibvafa Fallahpour<sup>1,2</sup>\*, Vincent Gureghian<sup>3</sup>\*, Guillaume J. Filion<sup>2</sup>, Ariel B. Lindner<sup>3</sup>, Amir Pandi<sup>3</sup>‡
|
65 |
+
|
66 |
+
<sup>1</sup> Vector Institute for Artificial Intelligence, Toronto ON, Canada
|
67 |
+
<sup>2</sup> University of Toronto Scarborough; Department of Biological Science; Scarborough ON, Canada
|
68 |
+
<sup>3</sup> Université Paris Cité, INSERM U1284, Center for Research and Interdisciplinarity, F-75006 Paris, France
|
69 |
+
\* These authors contributed equally to this work.
|
70 |
+
‡ To whom correspondence should be addressed: **amir.pandi@cri-paris.org** <br>
|
71 |
+
|
72 |
+
## Additional Resources
|
73 |
+
- **Project Website** <br>
|
74 |
+
https://adibvafa.github.io/CodonTransformer/
|
75 |
+
|
76 |
+
- **GitHub Repository** <br>
|
77 |
+
https://github.com/Adibvafa/CodonTransformer
|
78 |
+
|
79 |
+
- **Google Colab Demo** <br>
|
80 |
+
https://adibvafa.github.io/CodonTransformer/GoogleColab
|
81 |
+
|
82 |
+
- **PyPI Package** <br>
|
83 |
+
https://pypi.org/project/CodonTransformer/
|
84 |
+
|
85 |
+
- **Paper** <br>
|
86 |
+
TBD
|