Skip to content

Picker: PARSED

Jump to bottom
Seb Seager edited this page Apr 15, 2023 · 3 revisions

Summary

PARSED is available as a collection of Python programs. We used Python 3.6 for this installation.

The PARSED paper can be found here. Both the PARSED package files and a user manual can be downloaded in the parsed_v1.zip archive available with the supplementary materials of the PARSED paper. We include all relevant files in pickers/parsed/ of this repo for ease of access. In doing so, we adopt for these files the same usage license identified in the paper, namely:

the PARSED package and user manual [are available] for noncommercial use

Installation

First, make a new directory for the PARSED package.

mkdir parsed

PARSED is available as it was originally published in pickers/parsed/original/. Modified versions of these files with our patches and bug fixes are available in in pickers/parsed/. We will assume the patched files are being used, but the original software can be used at any time by replacing references to pickers/parsed/ with pickers/parsed/original where applicable.

Create and activate a new conda environment with the required dependencies. It may take conda several tries to solve this environment due to the highly specific package dependencies. The versions specified for each package correspond with those listed on the first page of the PARSED user manual. It may be possible to use later versions of some packages (e.g., mrcfile) without adverse effects.

conda create -n parsed -c conda-forge h5py=2.7.1 keras=2.0.8 numba=0.37.0 pandas=0.20.3 matplotlib=2.1.0 mrcfile=1.1.2 trackpy=0.4.1 tensorflow-gpu=1.7.0 pip python=3.6
conda activate parsed

Install any additional dependencies (not directly available through conda) with pip. In doing so, however, it is important that we use the pip executable that was just installed in the conda environment created above. To verify this, check that running which pip outputs something like /path/to/conda_envs/parsed/bin/pip. If not, try restarting your shell session or running conda deactivate parsed followed by conda activate parsed before proceeding.

pip install opencv-python==3.1.*

Usage

Overview

Inputs

  • micrographs for which to pick particles
  • picking aperture size, micrograph pixel resolution, and other configurable parameters
  • general model *.h5 file

Outputs (in a single directory)

  • *.star coordinate files
  • *.star extended parameter files (containing coordinates as well as blob mass and other data)
  • *.h5 data files (not used)

Pick using pretrained model

Start by collecting the micrograph files (*.mrc) to be picked in a directory (assuming they are not already available in their own directory). If you would like to use an existing public data set, our guide to the EMPIAR database may be helpful.

mkdir -p name_of_data_set/mrc
mv path/to/your_mrc_files/*.mrc name_of_data_set/mrc

Here we will use the micrographs located in demo_data/ as an example. Create another directory, in which any output, temporary, or configuration files will be saved by the picker.

mkdir demo_data/parsed_out

Use the following command to pick all micrographs in demo_data/mrc/. A description of parameters (modified from the PARSED user manual) is given below.

python -W ignore parsed_main.py --model=pickers/parsed/pre_train_model.h5 --data_path=demo_data/mrc/ --output_path=demo_data/parsed_out/ --file_pattern=*.mrc --job_suffix=autopick --angpixel=1.34 --img_size=4096 --edge_cut=0  --core_num=4 --aperture=160 --mass_min=4

When providing arguments for the command above, relative directories can be specified with ./ and ../, but do not use ~, as it does not appear to expand to the user's home folder. Note that if PARSED does not recognize a path, it may exit with the usual No such file or directory, or it may also attempt to pick anyway, returning Coordinates extraction finished 0 TotalPickNum found.

Parameters

  • --model: pre-trained segmentation model
  • --data_path: path to directory of input micrographs
  • --output_path: path to output directory
  • --file_pattern: regex (regular expression)-like pattern that describes the filenames of the micrographs to be picked (e.g., *.mrc includes only files ending in .mrc)
  • --job_suffix: suffix to be appended to the filenames of the output coordinate files (e.g., the input micrograph Falcon_2012_06_12-15_27_22_0.mrc would correspond with an output coordinate file named Falcon_2012_06_12-15_27_22_0_autopick.star)
  • --angpixel: sample rate (pixel resolution) for specified micrographs (in Å/pixel)
  • --img_size: either the width or height of the micrograph in pixels—whichever is larger
  • --edge_cut: number of pixels to "crop" from each edge of the micrograph
  • --core_num: number of processes to run in parallel
  • --aperture: diameter of the particle-picking aperture (in Å)
  • --mass_min: minimal picking mass of detected blobs (particle candidates)

Note: a GPU ID to use for picking can also be specified by appending the --gpu_id # flag, where # is the GPU ID. On an NVIDIA-based system, use the nvidia-smi command to get a list of available GPUs and their IDs.

Clone this wiki locally