Installation Fails

[DarioS]

I tried to install the software, but after about a minute of running $ conda env create -f SpecterEnv.yml it stops with error:

$ conda env create -f SpecterEnv.yml
Solving environment: failed

This seems to be caused by missing packages. I can't install those, either. For example,

$ conda install appnope
Solving environment: failed

PackagesNotFoundError: The following packages are not available from current channels:

  - appnope

Current channels:

  - https://repo.continuum.io/pkgs/main/linux-64
  - https://repo.continuum.io/pkgs/main/noarch
  - https://repo.continuum.io/pkgs/free/linux-64
  - https://repo.continuum.io/pkgs/free/noarch
  - https://repo.continuum.io/pkgs/r/linux-64
  - https://repo.continuum.io/pkgs/r/noarch
  - https://repo.continuum.io/pkgs/pro/linux-64
  - https://repo.continuum.io/pkgs/pro/noarch

The installation procedure is not as refined as for OpenMS, which I had no problem installing. Could you make it easier for people who have never used conda (or Python) before?

[rpeckner-broad]

Thanks for pointing this out. It seems that appnope is located on the anaconda channel, which wasn't included in the .yml. I've added it and hope it fixes the problem.

[DarioS]

Thanks. I showed just one example. The full list of unstated dependencies is:

Solving environment: failed

ResolvePackageNotFound: 
  - appscript==1.0.1=py27_0
  - pyaudio==0.2.7=py27_0
  - python.app==1.2=py27_4
  - xlwings==0.10.2=py27_0
  - appnope==0.1.0=py27_0
  - gettext==0.19.8=1
  - pandoc==1.19.2.1=0

I'm not sure why appnope still appears in the error message and it's a bit weird that there's a dependency on pyaudio.

[rpeckner-broad]

I've replaced SpecterEnv.yml with a clean export and verified that it works on an AWS VM. Could you check whether that fixes the problem for you?

[DarioS]

I am on Easter holidays and using my Windows laptop computer rather than the Ubuntu Linux server. I had more errors:

# Anaconda Prompt software
(base) C:\Users\Dario\Downloads\Specter-master>conda env create -f SpecterEnv.yml
Solving environment: failed

ResolvePackageNotFound:
  - libgfortran==3.0.0=1
  - tk==8.5.18=0
  - readline==6.2=2
  - mkl==11.3.3=0
  - cvxopt==1.1.8=py27_3
  - openssl==1.0.2j=0
  - zlib==1.2.8=3
  - sqlite==3.13.0=0

None of the packages are in common with the previous error message.

I also had another error installing pymzml:

(base) C:\Users\Dario\Downloads\Specter-master>conda install pymzml
Solving environment: failed

PackagesNotFoundError: The following packages are not available from current channels:

  - pymzml

Current channels:

  - https://repo.continuum.io/pkgs/main/win-32
  - https://repo.continuum.io/pkgs/main/noarch
  - https://repo.continuum.io/pkgs/free/win-32
  - https://repo.continuum.io/pkgs/free/noarch
  - https://repo.continuum.io/pkgs/r/win-32
  - https://repo.continuum.io/pkgs/r/noarch
  - https://repo.continuum.io/pkgs/pro/win-32
  - https://repo.continuum.io/pkgs/pro/noarch
  - https://repo.continuum.io/pkgs/msys2/win-32
  - https://repo.continuum.io/pkgs/msys2/noarch

Is there a channel specification missing for it?

[rpeckner-broad]

This looks like an OS problem - the yml was generated from a Linux machine using the linux-64 channels, which may contain packages not available from the win-32 channels. That said, I noticed some missing dependencies in the yml and have updated it accordingly, so you can give it a try when you have access to a Linux machine. This should also remove the need to install pymzml separately.

[DarioS]

The latest version's conda install works for me on the 64-bit Ubuntu Linux computer. However, can you provide more instructions for how to set PYSPARK_DRIVER_PYTHON and PYSPARK_PYTHON? The folders shown in the instructions aren't created on my computer and I do not know where the SpecterEnv folder has been made.

$ ls /usr/anaconda/envs/SpecterEnv/bin/python
ls: cannot access '/usr/anaconda/envs/SpecterEnv/bin/python': No such file or directory

There should be general information about where the folder is created.

[rpeckner-broad]

Unfortunately there's no general answer for this, as the location where anaconda stores its environments can vary widely from one user to the next. However, if you enter conda info -e at the command prompt you will get a list of your current environments along with their absolute paths. Look for "SpecterEnv" in this list and replace /usr/anaconda/envs/SpecterEnv in the code above with the path indicated there.

[DarioS]

It seems that the default is to put the environment into the user's home directory in an invisible folder:

# conda environments:
#
SpecterEnv               /home/dario/.conda/envs/SpecterEnv
base                  *  /home/dario/anaconda2

Would a hint in the Specter documentation to use conda info -e be useful to users who never used Anaconda before? I also installed OpenSWATH but found it straightforward.