title	README
author	Hugo MARTIN
date	09/01/2021

Contents

Overview of the Project
Describing the Workspace
Using Conda Environment
Launching Analysis
GitLab Link

Overview of the Project

The goal of the project was to study the efficiency of the neural network model to predict types of cancers
from the Cancer Genome Atlas Dataset (see report).

Prerequites

This project was designed on a Linux system, it is recommended to be on this Operating System to execute scripts.
Figures and output files can be read on all Operating Systems, but if you want execute scripts to make analysis,
it is better to be on a Linux system.

The execution of scripts need to activate the Conda Environment.
All the analysis were launched and all the figures were produced, so it is not necessary to launch analysis to generate and read figures.

Due to troubles to send data on GitHub, folders data and output were zipped, please unzip before to launch the project.

Describing the Workspace

Conda_environment/

• BiostatCondaEnv.yml

README.html

README.md

data/

• data.csv
• labels.csv

deleteAllOutput.sh

figures/

• AccuracyCurveWith3Layers100Epochs.eps
• AccuracyCurveWith3Layers100EpochsL1Equal0.01.eps
• AccuracyCurveWith3Layers100EpochsL1Equal0.1.eps
• AccuracyCurveWith3Layers20EpochsL1Equal0.01.eps
• AccuracyCurveWith3Layers20EpochsL1equal0.1.eps
• AccuracyCurveWith3Layers2OEpochs.eps
• AccuracyCurveWith3Layers50Epochs.eps
• AccuracyCurveWith3Layers50EpochsL1Equal0.01.eps
• AccuracyCurveWith3Layers50EpochsL1Equal0.1.eps
• AccuracyCurveWith4Layers100Epochs4Layers.eps
• AccuracyCurveWith4Layers100EpochsL1Equal0.01.eps
• AccuracyCurveWith4Layers100EpochsL1Equal0.1.eps
• AccuracyCurveWith4Layers20Epochs4Layers.eps
• AccuracyCurveWith4Layers20EpochsAndL1Equal0.01.eps
• AccuracyCurveWith4Layers20EpochsAndL1Equal0.1.eps
• AccuracyCurveWith4Layers50Epochs4Layers.eps
• AccuracyCurveWith4Layers50EpochsL1Equal0.01.eps
• AccuracyCurveWith4Layers50EpochsL1Equal0.1.eps
• BestNeuralNetworkAccuracyPlot.eps
• BestNeuralNetworkLossCurve.eps
• LearningCurveWithFourLayers.eps
• LearningCurveWithFourLayersAndL1Equal0.01.eps
• LearningCurveWithFourLayersAndL1Equal0.1.eps
• LearningCurveWithThreeLayers.eps
• LearningCurveWithThreeLayersAndL1Equal0.01.eps
• LearningCurveWithThreeLayersAndL1Equal0.1.eps
• PCAFigure.eps
• SupplementaryData.docx
• SupplementaryData.pdf
• UMAPFigure.eps

lib/

• AssamblingData.py
• NeuronalNetworkOnSubsetOfData.py
• PCAVisualisation.py
• SimpleNeuronalNetwork.py
• UMAPVisualisation.py
• VarianceGenes.R

main.py

output/

• 2264GenesDataframe/
  • newDataFrame.csv
• 399GenesDataframe/
  • newDataFrame.csv
• CompleteData.csv

report/

• Designing-a-Machine-learning-strategy-to-predict-types-of-cancers.pdf

---

Conda_environment contains the Conda Environment used to execute some codes.
data is the folder containing all the input data to generate output.
ouput is the folder containing all the output data.
figures is the folder containing all the figures generated from output.
report is the folder containing the report.
lib is the folder containing all the codes used to generate output data.
figures/SupplementaryData.pdf contains all the figures generated during the study.

Using Conda Environment

Installing Conda

If you didn't install Conda on your computer, you can follow this link to install it.

Deploying the Conda Environment on your computer

To deploy the Conda environment on your computer, go to the Conda_environment folder. Then, use the command:

conda env create -f BiostatCondaEnv.yml

To deploy the environment.

Activating the Conda Environment

To use the Conda environment on your computer, you must entry the command:

conda activate Biostat

MiniConda Environment Content

The used Conda packages for this project are listed below.

name: Biostat\

channels:

• conda-forge
• bioconda
• defaults
• r

dependencies:

• _libgcc_mutex=0.1=conda_forge
• _openmp_mutex=4.5=1_gnu
• _py-xgboost-mutex=2.0=cpu_0
• _r-mutex=1.0.1=anacondar_1
• absl-py=0.11.0=py36h5fab9bb_0
• astor=0.8.1=pyh9f0ad1d_0
• binutils_impl_linux-64=2.35.1=h193b22a_1
• binutils_linux-64=2.35=hc3fd857_29
• bwidget=1.9.14=ha770c72_0
• bzip2=1.0.8=h7f98852_4
• c-ares=1.17.1=h36c2ea0_0
• ca-certificates=2020.12.5=ha878542_0
• cached-property=1.5.1=py_0
• cairo=1.16.0=h9f066cc_1006
• certifi=2020.12.5=py36h5fab9bb_0
• curl=7.71.1=he644dc0_8
• cycler=0.10.0=py_2
• dbus=1.13.6=hfdff14a_1
• expat=2.2.9=he1b5a44_2
• fontconfig=2.13.1=h7e3eb15_1002
• freetype=2.10.4=h7ca028e_0
• fribidi=1.0.10=h36c2ea0_0
• gast=0.4.0=pyh9f0ad1d_0
• gcc_impl_linux-64=9.3.0=h28f5a38_17
• gcc_linux-64=9.3.0=h7247604_29
• gettext=0.19.8.1=h0b5b191_1005
• gfortran_impl_linux-64=9.3.0=h2bb4189_17
• gfortran_linux-64=9.3.0=ha1c937c_29
• glib=2.66.4=hcd2ae1e_1
• google-pasta=0.2.0=pyh8c360ce_0
• graphite2=1.3.13=h58526e2_1001
• grpcio=1.34.0=py36h8e87921_0
• gsl=2.6=he838d99_1
• gst-plugins-base=1.14.5=h0935bb2_2
• gstreamer=1.18.2=h3560a44_1
• gxx_impl_linux-64=9.3.0=h53cdd4c_17
• gxx_linux-64=9.3.0=h0d07fa4_29
• h5py=3.1.0=nompi_py36hc1bc4f5_100
• harfbuzz=2.7.2=ha5b49bf_1
• hdf5=1.10.6=nompi_h6a2412b_1113
• icu=67.1=he1b5a44_0
• importlib-metadata=3.3.0=py36h5fab9bb_2
• joblib=1.0.0=pyhd8ed1ab_0
• jpeg=9d=h36c2ea0_0
• keras=2.3.1=py36_0
• keras-applications=1.0.8=py_1
• keras-preprocessing=1.1.0=py_0
• kernel-headers_linux-64=2.6.32=h77966d4_13
• kiwisolver=1.3.1=py36h51d7077_0
• krb5=1.17.2=h926e7f8_0
• lcms2=2.11=hcbb858e_1
• ld_impl_linux-64=2.35.1=hea4e1c9_1
• libblas=3.9.0=6_openblas
• libcblas=3.9.0=6_openblas
• libclang=11.0.0=default_ha5c780c_2
• libcurl=7.71.1=hcdd3856_8
• libedit=3.1.20191231=he28a2e2_2
• libev=4.33=h516909a_1
• libevent=2.1.10=hcdb4288_3
• libffi=3.3=h58526e2_2
• libgcc-devel_linux-64=9.3.0=hfd08b2a_17
• libgcc-ng=9.3.0=h5dbcf3e_17
• libgfortran-ng=9.3.0=he4bcb1c_17
• libgfortran5=9.3.0=he4bcb1c_17
• libglib=2.66.4=h164308a_1
• libgomp=9.3.0=h5dbcf3e_17
• libgpuarray=0.7.6=h14c3975_1003
• libiconv=1.16=h516909a_0
• liblapack=3.9.0=6_openblas
• libllvm10=10.0.1=he513fc3_3
• libllvm11=11.0.0=he513fc3_0
• libnghttp2=1.41.0=h8cfc5f6_2
• libopenblas=0.3.12=pthreads_h4812303_1
• libpng=1.6.37=h21135ba_2
• libpq=12.3=h255efa7_3
• libprotobuf=3.14.0=h780b84a_0
• libssh2=1.9.0=hab1572f_5
• libstdcxx-devel_linux-64=9.3.0=h4084dd6_17
• libstdcxx-ng=9.3.0=h2ae2ef3_17
• libtiff=4.2.0=hdc55705_0
• libuuid=2.32.1=h7f98852_1000
• libwebp-base=1.1.0=h36c2ea0_3
• libxcb=1.13=h14c3975_1002
• libxgboost=1.3.0=h9c3ff4c_1
• libxkbcommon=1.0.3=he3ba5ed_0
• libxml2=2.9.10=h68273f3_2
• llvmlite=0.35.0=py36h05121d2_0
• lz4-c=1.9.3=h9c3ff4c_0
• make=4.3=hd18ef5c_1
• mako=1.1.3=pyh9f0ad1d_0
• markdown=3.3.3=pyh9f0ad1d_0
• markupsafe=1.1.1=py36he6145b8_2
• matplotlib=3.3.3=py36h5fab9bb_0
• matplotlib-base=3.3.3=py36he12231b_0
• mysql-common=8.0.22=ha770c72_1
• mysql-libs=8.0.22=h1fd7589_1
• ncurses=6.2=h58526e2_4
• nspr=4.29=he1b5a44_1
• nss=3.60=hb5efdd6_0
• numba=0.52.0=py36h284efc9_0
• numpy=1.19.4=py36h2aa4a07_2
• olefile=0.46=pyh9f0ad1d_1
• openssl=1.1.1i=h7f98852_0
• pandas=1.1.5=py36h284efc9_0
• pandoc=2.11.3.2=h7f98852_0
• pango=1.42.4=h69149e4_5
• pcre=8.44=he1b5a44_0
• pcre2=10.36=h032f7d1_0
• pillow=8.1.0=py36h4f9996e_0
• pip=20.3.3=pyhd8ed1ab_0
• pixman=0.40.0=h36c2ea0_0
• protobuf=3.14.0=py36hc4f0c31_0
• pthread-stubs=0.4=h36c2ea0_1001
• py-xgboost=1.3.0=py36h5fab9bb_1
• pygpu=0.7.6=py36h68bb277_1002
• pyparsing=2.4.7=pyh9f0ad1d_0
• pyqt=5.12.3=py36h5fab9bb_6
• pyqt-impl=5.12.3=py36h7ec31b9_6
• pyqt5-sip=4.19.18=py36hc4f0c31_6
• pyqtchart=5.12=py36h7ec31b9_6
• pyqtwebengine=5.12.1=py36h7ec31b9_6
• python=3.6.12=hffdb5ce_0_cpython
• python-dateutil=2.8.1=py_0
• python_abi=3.6=1_cp36m
• pytz=2020.5=pyhd8ed1ab_0
• pyyaml=5.3.1=py36he6145b8_1
• qt=5.12.9=h763d07f_1
• r-base=4.0.3=ha43b4e8_3
• r-base64enc=0.1_3=r40hcdcec82_1004
• r-digest=0.6.27=r40h1b71b39_0
• r-evaluate=0.14=r40h6115d3f_2
• r-glue=1.4.2=r40hcdcec82_0
• r-highr=0.8=r40h6115d3f_2
• r-htmltools=0.5.0=r40h0357c0b_0
• r-jsonlite=1.7.2=r40hcfec24a_0
• r-knitr=1.30=r40h6115d3f_0
• r-lattice=0.20_41=r40hcdcec82_2
• r-magrittr=2.0.1=r40h9e2df91_1
• r-markdown=1.1=r40hcdcec82_1
• r-matrix=1.3_0=r40he454529_0
• r-mime=0.9=r40hcdcec82_1
• r-rappdirs=0.3.1=r40hcdcec82_1004
• r-rcpp=1.0.5=r40he524a50_0
• r-reticulate=1.18=r40h1b71b39_0
• r-rlang=0.4.10=r40hcfec24a_0
• r-rmarkdown=2.6=r40hc72bb7e_0
• r-stringi=1.5.3=r40h604b29c_0
• r-stringr=1.4.0=r40h6115d3f_2
• r-tinytex=0.28=r40hc72bb7e_0
• r-xfun=0.19=r40h9e2df91_0
• r-yaml=2.2.1=r40hcdcec82_1
• readline=8.0=he28a2e2_2
• scikit-learn=0.24.0=py36he4fde30_0
• scipy=1.5.3=py36h9e8f40b_0
• sed=4.8=he412f7d_0
• setuptools=49.6.0=py36h9880bd3_2
• six=1.15.0=pyh9f0ad1d_0
• sqlite=3.34.0=h74cdb3f_0
• sysroot_linux-64=2.12=h77966d4_13
• tbb=2020.2=h4bd325d_2
• tensorboard=1.14.0=py36_0
• tensorflow=1.14.0=hc3e5e64_0
• tensorflow-base=1.14.0=py36hc3e5e64_0
• tensorflow-estimator=1.14.0=py36h5ca1d4c_0
• termcolor=1.1.0=py_2
• theano=0.9.0=py36_1
• threadpoolctl=2.1.0=pyh5ca1d4c_0
• tk=8.6.10=h21135ba_1
• tktable=2.10=hb7b940f_3
• tornado=6.1=py36h1d69622_0
• typing_extensions=3.7.4.3=py_0
• umap-learn=0.4.6=py36h9f0ad1d_0
• werkzeug=1.0.1=pyh9f0ad1d_0
• wheel=0.36.2=pyhd3deb0d_0
• wrapt=1.12.1=py36h1d69622_2
• xgboost=1.3.0=py36hc4f0c31_1
• xorg-kbproto=1.0.7=h14c3975_1002
• xorg-libice=1.0.10=h516909a_0
• xorg-libsm=1.2.3=h84519dc_1000
• xorg-libx11=1.6.12=h516909a_0
• xorg-libxau=1.0.9=h14c3975_0
• xorg-libxdmcp=1.1.3=h516909a_0
• xorg-libxext=1.3.4=h516909a_0
• xorg-libxrender=0.9.10=h516909a_1002
• xorg-libxt=1.2.0=h516909a_0
• xorg-renderproto=0.11.1=h14c3975_1002
• xorg-xextproto=7.3.0=h14c3975_1002
• xorg-xproto=7.0.31=h7f98852_1007
• xz=5.2.5=h516909a_1
• yaml=0.2.5=h516909a_0
• zipp=3.4.0=py_0
• zlib=1.2.11=h516909a_1010
• zstd=1.4.8=ha95c52a_1

prefix: /home/hugo/miniconda3/envs/Biostat

Launching Analysis

To generate all the output files and all the figures, you must execute the main.py python script.
For that, you must go in the root folder of the project and execute this command after activating the Conda Environement:

python3 main.py

To delete all the figures and the output files, you can execute the deleteAllOutput.sh bash script in the root folder of the project.
For that, you must execute this command:

./deleteAllOutput.sh

GitLab Link

This project is available on GitLab