biological sequence similarity analysing using Diabolo.
Software Requirements:
- Python3
- virtualenv or Anaconda
- CUDA 10.0 (Optional If using GPU)
- cuDNN (>= 7.4.1) (Optional If using GPU)
BioSeq-Diabolo has been tested on Windows, Ubuntu 16.04, and 18.04 operating systems.
virtualenv -p python3.7 venv
source ./venv/bin/activate
# You can use 'requirements' file:
pip install -r requirements.txt
# or directly install the corresponding package:
pip install matchzoo-py
pip install scikit-learn
pip install lightgbm
pip install seabornconda create -n venv python=3.7
conda activate venv
# You can use 'requirements' file:
pip install -r requirements.txt
# or directly install the corresponding package:
pip install matchzoo-py
pip install scikit-learn
pip install lightgbm
pip install seabornpython sesica_clf.py -base_dir ../data/web_hetero -data_type hetero -bmk_vec_a ../data/web_hetero/bmk_vec_a.txt -bmk_vec_b ../data/web_hetero/bmk_vec_b.txt -bmk_label ../data/web_hetero/bmk_pos_label.txt ../data/web_hetero/bmk_neg_label.txt -ind score -ind_vec_a ../data/web_hetero/ind_vec_a.txt -clf svm rf ert knn mnb gbdt goss dart mlp -metric aupr
python sesica_rank.py -base_dir ../data/web_hetero -rank ltr -clf svm rf ert knn mnb gbdt goss dart mlp -rank ltr -metric aupr
python sesica_plot.py -base_dir ../data/web_hetero -data_type hetero -clf svm rf ert knn mnb gbdt goss dart mlp -plot pie net roc prc box dist dr hppython sesica_clf.py -base_dir ../data/iCircDA -data_type hetero -bmk_vec_a ../data/iCircDA/bmk_circRNA.txt -bmk_vec_b ../data/iCircDA/bmk_disease.txt -bmk_label ../data/iCircDA/benchmark_pos.txt ../data/iCircDA/benchmark_neg.txt -clf svm rf ert knn mnb gbdt goss mlp -metric auc -gs_mode 2
python sesica_rank.py -base_dir ../data/iCircDA -rank ltr -clf svm rf knn mnb goss -metric auc -gs_mode 2
python sesica_plot.py -base_dir ../data/iCircDA -data_type hetero -clf svm rf knn mnb goss -rank ltr -plot roc polar hp dr pie -plot_set testpython sesica_clf.py -base_dir ../data/ProtRe -data_type homo -bmk_vec ../data/ProtRe/bmk_vec.txt -bmk_label ../data/ProtRe/pos_label.txt ../data/ProtRe/neg_label.txt -clf svm rf ert knn gbdt goss dart mlp -metric roc@1 -gs_mode 2
python sesica_rank.py -base_dir ../data/ProtRe -rank ltr -clf svm rf ert knn mlp -metric roc@1 -gs_mode 2
python sesica_plot.py -base_dir ../data/ProtRe -data_type homo -clf svm rf ert knn mlp -rank ltr -plot roc prc box polar hp dr dist pie bar -plot_set testpython sesica_clf.py -base_dir ../data/go -data_type hetero -bmk_vec_a ../data/go/cc_bmk_vec_a.txt -bmk_vec_b ../data/go/cc_bmk_vec_b.txt -bmk_label ../data/go/pos_label.txt ../data/go/neg_label.txt -clf svm rf ert knn mnb gbdt goss dart mlp -metric aupr -gs_mode 2
python sesica_rank.py -base_dir ../data/go -rank ltr -clf svm rf ert knn mlp -metric aupr -gs_mode 2
python sesica_plot.py -base_dir ../data/go -data_type homo -clf svm rf ert knn mlp -rank ltr -plot polar dr dist pie bar -plot_set testdeep-learning semantic similarity calculation reference
LTR part code reference
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent