Examples¶

Example analyses
Running the examples

Example analyses ¶

Each example analysis is a published study by the Bloom lab. Each analysis is in its own subdirectory in the examples folder on GitHub. Typically there is a Jupyter notebook that performs the analysis.

Here are the analyses:

Doud and Bloom (2016)¶

Summary: Deep mutational scanning of influenza hemagglutinin using Barcoded-subamplicon sequencing, then estimation of the Amino-acid preferences.

Citation: Doud and Bloom (2016)

Jupyter notebook: Doud2016 example

Doud et al (2017)¶

Summary: Mutational antigenic profiling of influenza hemagglutinin against strain-specific antibodies using Barcoded-subamplicon sequencing, then estimation of the Differential selection from each antibody

Citation: Doud et al (2017)

Jupyter notebook:Doud2017 example.

Dingens et al (2017)¶

Summary: Mutational antigenic profiling of HIV Env (BF520 strain) against PGT151 antibody using Barcoded-subamplicon sequencing, and then estimation of the Differential selection.

Citation: Dingens et al (2017)

Jupyter notebook: Dingens2017 example.

Doud et al (2018)¶

Summary: Mutational antigenic profiling of influenza hemagglutinin against broadly neutralizing and strain-specific antibodies using Barcoded-subamplicon sequencing, then estimation of the Fraction surviving.

Citation: Doud et al (2018)

Jupyter notebook: Doud2018 example.

Haddox et al (2018)¶

Summary: Deep mutational scanning of two HIV Envs using Barcoded-subamplicon sequencing, then comparison of the Amino-acid preferences.

Citation: Haddox et al (2018)

Jupyter notebook: Haddox2018 example.

Dingens et al (2018)¶

Summary: Mutational antigenic profiling of HIV Env against fusion-peptide antibodies usring Barcoded-subamplicon sequencing, then estimation of the Differential selection.

Citations: Dingens et al (2018)

Jupyter notebook: Dingens2018 example

Lee et al (2018)¶

Summary: Deep mutational scanning of H3 influenza hemagglutinin using Barcoded-subamplicon sequencing, then estimation of the Amino-acid preferences.

Citations: Lee et al (2018)

Jupyter notebook: Lee2018 example

Running the examples ¶

Navigate here to access the example analyses and require input data.

To run a Jupyter notebook, you need to install jupyter. Most of the notebooks also have other requirements (such as fastq-dump to download the deep sequencing data from the Sequence Read Archive), and these dependencies are described in the notebooks.

To run a notebook interactively, type:

jupyter notebook analysis_notebook.ipynb

To run a notebook in command-line mode, run:

jupyter nbconvert --to notebook --execute --inplace --ExecutePreprocessor.timeout=-1 analysis_notebook.ipynb

You should not open the notebook while it is being run in command-line mode.

If you are in the Bloom lab and using the Hutch server, you can run a notebook via slurm by creating a file run.sbatch that looks like this:

#!/bin/sh
#SBATCH
#PBS -l walltime=24:00:00
jupyter nbconvert --to notebook --execute --inplace --ExecutePreprocessor.timeout=-1 analysis_notebook.ipynb

and then running the file with:

sbatch -p largenode -c 14 --mem=140000 run.sbatch

where the number of CPUs requested with -c should match the number specified for multi-processor operations in the notebook.