blobulator 0.1.2

Edge Detection in Protein Sequences

Protein Blobulator

Looking for the web interface? Find it here: https://www.blobulator.branniganlab.org/

This tool identifies contiguous stretches of hydrophobic residues within a protein sequence. Any sequence of contiguous hydrophobic residues that is at least as long as the minimum blob length is considered an hydrophobic or h "blob". Any remaining segments that are at least as long as the minimum length are considered polar or p "blobs," while those that are shorter than the minimum blob length are considered separator or "s" residues. Separator residues are very short stretches of non-hydrophobic residues that may be found between two h blobs.

Running locally:

Installation guide:

Software requirements:

Python 3.9+

Quick Install:

[Optional] Create a conda environment:

conda create --name blobulator_env python=3.9
conda activate blobulator_env

[For website and sample scripts] Download the repository:

git clone https://github.com/BranniganLab/blobulator

Install with pip

pip install git+https://github.com/BranniganLab/blobulator

Known issue: If you get an error installing pycairo, try conda install pycairo and retry the above.

Running through an internet browser:

Note: this option is identical to the website version, but is hosted on your local machine:

cd [path_to_repository]/website
python3 blobulation.py

If a browser doesn't open automatically, copy the url from the terminal into a browser.

Scripting - Hello, World:

    import blobulator

    # A very simple oligopeptide and standard settings
    sequence = "RRRRRRRRRIIIIIIIII"
    cutoff = 0.4
    min_blob = 4
    hscale = "kyte_doolittle"

    # Do the blobulation
    blobDF = blobulator.compute(sequence, cutoff, min_blob, hscale)
    
    # Cleanup the dataframe (make it more human-readable)
    blobDF = blobulator.clean_df(blobDF)
    
    # Save it as a csv for later use
    oname = "hello_blob.csv"
    blobDF.to_csv(oname, index=False)

Additional sample scripts can be found in the repository examples directory.

Using the command-line utility blobulate.py:

Basic usage:

Open a terminal in the blobulator directory and run:

python3 [path_to_repository]/examples/blobulate.py --sequence AFRPGAGQPPRRKECTPEVEEGV --oname ./my_blobulation.csv

This will blobulate the sequence "AFRPGAGQPPRRKECTPEVEEGV" and write the result to my_blobulation.csv

Options:

You may specify additional paramters using the following options:

-h, --help           show help information and exit

--sequence SEQUENCE  Takes a single string of EITHER DNA or protein one-letter codes (no spaces).
--cutoff CUTOFF      Sets the cutoff hydrophobicity (floating point number between 0.00 and 1.00 inclusive). Defaults to 0.4
--minBlob MINBLOB    Mininmum blob length (integer greater than 1). Defaults to 4
--oname ONAME        Name of output file or path to output directory. Defaults to blobulated_.csv
--fasta FASTA        FASTA file with 1 or more sequences
--DNA DNA            Flag that says whether the inputs are DNA or protein. Defaults to false (protein)

Advanced Usage (FASTA files):

• Place a fasta file with one or more sequences in any directory (Note: they must all be DNA or protein sequences)
• Open a terminal in the blobulator directory and run:

python3 [path_to_repository]/examples/blobulate.py --fasta ./relative/path/to/my_sequences.fasta --oname ./relative/path/to/outputs/

• This will blobulate all sequences in my_sequences.fasta (assuming they are protein sequences) and output the results to the outputs folder prefixed by their sequence id.

Example:

There is a fasta file in blobulation/example called b_subtilis.fasta that contains the sequences of several proteins from Bacillus subtilis. To blobulate all those proteins with a cutoff of 0.4 and a minimum blob size of 4, we run:

mkdir outputs
python3 [path_to_repository]/examples/blobulate.py --fasta ../example/b_subtilis.fasta --cutoff 0.4 --minBlob 4 --oname outputs/

CSV Outputs:

Whether you have blobulated your proteins of interest using the web utility or the command-line option, you can obtain the blobulation data as a csv (the only output of the command line option or by clicking "Download Data" on the website). These CSVs are organized with each residue in its own row and columns as follows:

• Residue_Number: The position of the residue in the sequence starting at 1
• Residue_Name: The one-letter amino acid code
• Window: The size of the rolling average window (this is currently 3 by default. We have not yet added the ability to change this.)
• Hydropathy_Cutoff: The normalized cutoff used during blobulation (float between 0 and 1)
• Minimum_Blob_Length: The minimum blob length used during blobulation (integer greater than 0)
• blob_length: The length of the residue's blob
• Normalized_Mean_Blob_Hydropathy: The normalized mean hydropathy of the residue's blob
• Blob_Type: The one-letter blob code (h=hydropathic, p=polar/hydrophilic, s=short hydrophilic)
• Blob_Index_Number: Indices which distinguish blobs. E.g. h1 is the first hydrophobic blob. h1a and h1b refer to two halves of a blob separated by a short hydrophobic blob.
• Blob_Das-Pappu_Class: Blob scored by Das-Pappu globularity. 1=globular, 2=Janus/boundary, 3=Polar, 4=Polycation, 5=Polyanion
• Blob_NCPR: Net-charge-per-residue of the blob
• Fraction_of_Positively_Charged_Residues: FPC = N(Positively charged residues)/N(residues)
• Fraction_of_Negatively_Charged_Residues: FNC = N(Negatively charged residues)/N(residues)
• Fraction_of_Charged_Residues: FCR = FPC+FNC
• Uversky_Diagram_Score: Distance from the Uversky-Gillespie-Fink globular/disordered cutoff. See https://pubmed.ncbi.nlm.nih.gov/11025552/
• dSNP_enrichment: Predicted disease-causing mutation enrichment. dSNP_enrichment: Predicted enrichment of disease-causing SNPs. See Lohia, Hansen, and Brannigan, 2022, PNAS, In Press.
• Blob_Disorder: Mean expected disorder score as provided by D2P2. See https://doi.org/10.1093/nar/gks1226
• Normalized_Kyte-Doolittle_hydropathy: K-D hydropathy normalized to be between 0 and 1. See Kyte-Doolittle_hydropathy.
• Kyte-Doolittle_hydropathy: Traditional K-D hydropathy (on a scale from -4.5 to 4.5). This is a very common hydrophobicity scale dating to 1982: https://doi.org/10.1016%2F0022-2836%2882%2990515-0

Visualizing in VMD

There is a tcl script in the VMD_scripts directory that will read a csv from the website or the local tool.

To use it:

1. Load your protein of choice into a vmd session and open the tkconsole
2. Source this file using:
3. source /path/to/bctool.tcl
4. Get the protein sequence using:

set protSel [atomselect top "protein"]
get_sequence $protSel

6. Copy and paste the sequence into the blobulator and blobulate according to your needs
7. Download the data using the "Download Data" button on the website
8. Copy the csv to your working directory
9. Import the blobulation data using:

getBlobs my_blobulation.csv $protSel

11. Visualize according to your needs.

• User will contain 1=hydrophobic blob, 2=polar blob, 3=short blob
• User2 will contain the blob id number