
# About DeepDTA: deep drug-target binding affinity prediction

The approach used in this work is the modeling of protein sequences and compound 1D representations (SMILES) with convolutional neural networks (CNNs) to predict the binding affinity value of drug-target pairs.

![Figure](https://github.com/hkmztrk/DeepDTA/blob/master/docs/figures/deepdta.png?raw=true)
# Installation

## Data

Please see the [README](https://github.com/hkmztrk/DeepDTA/blob/master/data/README.md) for detailed explanation.

## Requirements

You'll need to install following in order to run the codes. Refer to [deepdta.yml](https://github.com/hkmztrk/DeepDTA/blob/master/deepdta.yml) for a conda environment tested in Linux. 

*  [Python 3.4 <=](https://www.python.org/downloads/)
*  [Keras 2.x](https://pypi.org/project/Keras/)
*  [Tensorflow 1.x](https://www.tensorflow.org/install/)
*  numpy
*  matplotlib
*  scikit-learn



You have to place "data" folder under "source" directory. 

# Usage
```
python run_experiments.py --num_windows 32 \
                          --seq_window_lengths 8 12 \
                          --smi_window_lengths 4 8 \
                          --batch_size 256 \
                          --num_epoch 100 \
                          --max_seq_len 1000 \
                          --max_smi_len 100 \
                          --dataset_path 'data/kiba/' \
                          --problem_type 1 \
                          --log_dir 'logs/'


```





**For citation:**

```
@article{ozturk2018deepdta,
  title={DeepDTA: deep drug--target binding affinity prediction},
  author={{\"O}zt{\"u}rk, Hakime and {\"O}zg{\"u}r, Arzucan and Ozkirimli, Elif},
  journal={Bioinformatics},
  volume={34},
  number={17},
  pages={i821--i829},
  year={2018},
  publisher={Oxford University Press}
}
```

	a/README.md		b/README.md
1	# About DeepDTA: deep drug-target binding affinity prediction	1	# About DeepDTA: deep drug-target binding affinity prediction
2		2
3	The approach used in this work is the modeling of protein sequences and compound 1D representations (SMILES) with convolutional neural networks (CNNs) to predict the binding affinity value of drug-target pairs.	3	The approach used in this work is the modeling of protein sequences and compound 1D representations (SMILES) with convolutional neural networks (CNNs) to predict the binding affinity value of drug-target pairs.
4		4
5	![Figure](https://github.com/hkmztrk/DeepDTA/blob/master/docs/figures/deepdta.PNG)	5	![Figure](https://github.com/hkmztrk/DeepDTA/blob/master/docs/figures/deepdta.png?raw=true)
6	# Installation	6	# Installation
7		7
8	## Data	8	## Data
9		9
10	Please see the [README](https://github.com/hkmztrk/DeepDTA/blob/master/data/README.md) for detailed explanation.	10	Please see the [README](https://github.com/hkmztrk/DeepDTA/blob/master/data/README.md) for detailed explanation.
11		11
12	## Requirements	12	## Requirements
13		13
14	You'll need to install following in order to run the codes. Refer to [deepdta.yml](https://github.com/hkmztrk/DeepDTA/blob/master/deepdta.yml) for a conda environment tested in Linux.	14	You'll need to install following in order to run the codes. Refer to [deepdta.yml](https://github.com/hkmztrk/DeepDTA/blob/master/deepdta.yml) for a conda environment tested in Linux.
15		15
16	* [Python 3.4 <=](https://www.python.org/downloads/)	16	* [Python 3.4 <=](https://www.python.org/downloads/)
17	* [Keras 2.x](https://pypi.org/project/Keras/)	17	* [Keras 2.x](https://pypi.org/project/Keras/)
18	* [Tensorflow 1.x](https://www.tensorflow.org/install/)	18	* [Tensorflow 1.x](https://www.tensorflow.org/install/)
19	* numpy	19	* numpy
20	* matplotlib	20	* matplotlib
21	* scikit-learn	21	* scikit-learn
22		22
23		23
24		24
25	You have to place "data" folder under "source" directory.	25	You have to place "data" folder under "source" directory.
26		26
27	# Usage	27	# Usage
28	```	28	```
29	python run_experiments.py --num_windows 32 \	29	python run_experiments.py --num_windows 32 \
30	--seq_window_lengths 8 12 \	30	--seq_window_lengths 8 12 \
31	--smi_window_lengths 4 8 \	31	--smi_window_lengths 4 8 \
32	--batch_size 256 \	32	--batch_size 256 \
33	--num_epoch 100 \	33	--num_epoch 100 \
34	--max_seq_len 1000 \	34	--max_seq_len 1000 \
35	--max_smi_len 100 \	35	--max_smi_len 100 \
36	--dataset_path 'data/kiba/' \	36	--dataset_path 'data/kiba/' \
37	--problem_type 1 \	37	--problem_type 1 \
38	--log_dir 'logs/'	38	--log_dir 'logs/'
39		39
40		40
41	```	41	```
42		42
43		43
44		44
45		45
46		46
47	For citation:	47	For citation:
48		48
49	```	49	```
50	@article{ozturk2018deepdta,	50	@article{ozturk2018deepdta,
51	title={DeepDTA: deep drug--target binding affinity prediction},	51	title={DeepDTA: deep drug--target binding affinity prediction},
52	author={{\"O}zt{\"u}rk, Hakime and {\"O}zg{\"u}r, Arzucan and Ozkirimli, Elif},	52	author={{\"O}zt{\"u}rk, Hakime and {\"O}zg{\"u}r, Arzucan and Ozkirimli, Elif},
53	journal={Bioinformatics},	53	journal={Bioinformatics},
54	volume={34},	54	volume={34},
55	number={17},	55	number={17},
56	pages={i821--i829},	56	pages={i821--i829},
57	year={2018},	57	year={2018},
58	publisher={Oxford University Press}	58	publisher={Oxford University Press}
59	}	59	}
60	```	60	```