otherprojectsforks / deeppep Goto Github PK

DeepPep, is a protein identification software which uses deep-convolutional neural network to predict the protein set from a proteomics mixture, given the sequence universe of possible proteins and a target peptide profile.

• torch7
• luarocks install cephes
• luarocks install csv
• SparseNN
• python3.4 or above
• biopython

git clone https://github.com/ameenetemady/MyCommon.git
git clone https://github.com/DeepPep/DeepPep.git

• Step1: prepare a directory containing your input files (with exact names):

  • identification.tsv: tab-delimeted file: column1: peptide, column2: protein name, column3: identification probability
  • db.fasta: reference protein database in fasta format.

• Step2: python run.py directoryName

Upon completion, pred.csv will contain the predicted protein identification probabilities.

There are 7 example datasets (used for benchmarking in the paper). Each dataset is generated from MS/MS raw files using TPP pipeline. For example, to run the 18Mix benchmark dataset, simply run the following:

python run.py data/18Mix

If you have any questions about DeepPep, please contact Minseung Kim ([email protected]) or Ameen Eetemadi ([email protected]).

M. Kim, A. Eetemadi, and I. Tagkopoulos, “DeepPep: deep proteome inference from peptide profiling”, PLoS Computational Biology (2017) [link]

See the LICENSE file for license rights and limitations (Apache2.0).

This work was supported by a grant from Mars, Inc. and NSF award 1516695.