Skip to content

Latest commit

 

History

History
390 lines (307 loc) · 21.9 KB

README.md

File metadata and controls

390 lines (307 loc) · 21.9 KB
Raw

CROPSR: An Automated Platform for Complex Genome-Wide CRISPR gRNA Design and Validation

About CROPSR

This repository is open sourced as specified in the LICENSE file. It is Apache License 2.0. For additional information, please check LICENSE.
CROPSR is a python tool designed for genome-wide gRNA design and evaluation for CRISPR experiments, with special focus on complex genomes such as those found in energy-producing crops. CROPSR is a product of the DOE Center for Advanced Bioenergy and Bioproducts Innovation (CABBI).

Citation

Please cite the following when utilizing CROPSR:

Müller Paul, H., Istanto, D.D., Heldenbrand, J. et al. CROPSR: an automated platform for complex genome-wide CRISPR gRNA design and validation. BMC Bioinformatics 23, 74 (2022). https://doi.org/10.1186/s12859-022-04593-2.

Table of Contents

About

Prerequisites

Getting Started

Example Data and Output

Disclosures

Prerequisites

CROPSR does not require a separate Python environment for dependency management, as there are few dependencies and the code is updated to their current versions. Any required changes will be made to maintain compatibility with current versions of these dependencies. CROPSR is intended to be used on Python 3.7 or newer (newest stable recommended).

Dependencies:

  • Python version: 3.7 or newer
  • Python libraries:
    • pandas
    • argparse

Installing dependencies:

  • If you have both Python 2.7 and Python 3 installed (e.g. Ubuntu 18.10 or older, MacOS Catalina or older), use pip3 to install the libraries to the correct path for Python 3:
$ pip3 install pandas argparse
  • If you have only Python 3 installed, or Python 3 is your default (e.g. Ubuntu 20.04 or newer, MacOS Big Sur or newer), the default instalation with :
$ pip install pandas argparse

Back to top

Geting Started

Input files

To perform a full genome analysis, the following two files are required:

  • Fasta file containing whole genome sequence
  • GFF file containing genome functional annotation

Imoprtant note for genomes downloaded from Phytozome:

If your genome comes from Phytozome, please make sure to also download the annotation_info.txt file. This is important because Phytozome GFF files contain a reference for a location within a separate file, so their genome browser can display the functional annotation as a separate layer. Without this file, CROPSR will output a database with no functional annotation when processing a Phytozome genome.

First steps

CROPSR was developed as a CLI software, and requires a basic understanding of bash (or equivalent).

  1. Download the project folder.

    $ git clone git@github.com:cabbi-bio/CROPSR.git

  2. Navigate to the project folder.

    $ cd CROPSR

  3. Place all input files on a folder (This does not need to be the CROPSR folder, as long as all input files are in the same folder. This is especially important for genomes downloaded from Phytozome).

  4. Running CROPSR (if Python 3 is set as your default python path, replace where it says python3): You can get the CROPSR help prompt by entering python3 CROPSR.py -h will return the list of arguments below:

    $ python3 CROPSR.py -h

    usage: CROPSR.py [-h] -f F -g G -p [-o O] [-l] [-L] [--cas9] [-v]

optional arguments:
  -h, --help        show this help message and exit
  -f , --fasta F   [required] path to input file in FASTA format
  -g , --gff G     path to input file in GFF format
  -p , --phytozome   path to input annotation info file in TXT format, default = None
  -o , --output O  path to output file, default = data.csv
  -l , --length     length of the gRNA sequence, default = 20
  -L , --flanking   length of flanking region for verification, default = 200
  --cas9            specifies that design will be made for the Cas9 CRISPR system
  -v, --verbose     prints visual indicators for each iteration

    CROPSR arguments

    Flag Description
    -h, --help Quits the code and opens the help prompt
    -f, --fasta Path to the FASTA file (*.fasta, *.fa) containing the genome sequence (always required)
    -g, --gff Path to the GFF file (*.gff, *.gff3) containing the functional annotation (always required)
    -p, --phytozome Path to the annotation_info.txt file containing functional annotation (required for phytozome genomes)
    -o, --output Path to save the output database file, including file name. The default is data.csv at the working directory
    -l, --length Desired length of the gRNA sequence. The default value is 20, and this should not be changed unless required by a non-standard Cas protein. Changing this value otherwise may cause the experiment to fail
    -L, --flanking Desired length of flanking region for designing primers for PCR validation. The default value is 200 bases upstream and downstream of the cutsite
    --cas9 Type of CRISPR system for the experiment. At least one CRISPR system is required. (Currently only Cas9 is available, but other systems may be implemented in a future version)
    -v, --verbose Enables verbose mode and prints notes at several points of the process. Enable for debugging

Output files

After completion, if verbose is enabled, a prompt will appear to inform the user that The output file has been generated at example_data/cropsr_output.csv. No temporary files are generated during the analysis.

CROPSR outputs a CSV (comma separated values) file by default. This file type was chosen due to the ease of handling, including importing it into the database manager of the user's preference. An option to output as a JSON following MongoDB formatting is also provided, requiring the pymongo library as an additional dependency.

Click here for MongoDB dependency instructions
  • If you have both Python 2.7 and Python 3 installed (e.g. Ubuntu 18.10 or older, MacOS Catalina or older), use pip3 to install the libraries to the correct path for Python 3:
$ pip3 install pymongo
  • If you have only Python 3 installed, or Python 3 is your default (e.g. Ubuntu 20.04 or newer, MacOS Big Sur or newer), the default instalation with :
$ pip install pymongo

Invalid/Unactionable gRNA Sequences

In the case that an invalid/unactionable gRNA sequence is generated from the genome FASTA file (i.e. possibly due to sequencing inaccuracies) the sequence will be stored in the output file with an "on_site_score" of -1, such that these sequences can be referenced by the user if desired, but will be excluded from the candidate gRNA database if the user queries limiting (0 ≤ on_site_score ≤ 1).

Back to top

Example data and output

An example data set is provided to serve as a tutorial. This data set is comprised of the first chromosome of Saccharomyces cerevisiae, and includes both a FASTA and GFF files. The entire process will be described below.

The example data set is provided in a folder named sample_data within the contents of this Git repository. To follow along with this tutorial, no additional data should be required.

The data structure of the repository is represented below:

+-- README.md
+-- LICENCE.md
+-- CROPSR.py
+-- cropsr_functions.py
+-- prmrdsgn2.py
+-- sample_data/
    +-- sample_genome.fa
    +-- sample_genome.gff
+-- .DS_Store
+-- .gitignore
Click here to preview FASTA file sample_genome.fa 
>Chr01
ccacaccacacccacacacccacacaccacaccacacaccacaccacacccacacacacacatCCTAACACTACCCTAAC
ACAGCCCTAATCTAACCCTGGCCAACCTGTCTCTCAACTTACCCTCCATTACCCTGCCTCCACTCGTTACCCTGTCCCAT
TCAACCATACCACTCCGAACCACCATCCATCCCTCTACTTACTACCACTCACCCACCGTTACCCTCCAATTACCCATATC
CAACCCACTGCCACTTACCCTACCATTACCCTACCATCCACCATGACCTACTCACCATACTGTTCTTCTACCCACCATAT
TGAAACGCTAACAAATGATCGTAAATAACACACACGTGCTTACCCTACCACTTTATACCACCACCACATGCCATACTCAC
CCTCACTTGTATACTGATTTTACGTACGCACACGGATGCTACAGTATATACCATCTCAAACTTACCCTACTCTCAGATTC
CACTTCACTCCATGGCCCATCTCTCACTGAATCAGTACCAAATGCACTCACATCATTATGCACGGCACTTGCCTCAGCGG
TCTATACCCTGTGCCATTTACCCATAACGCCCATCATTATCCACATTTTGATATCTATATCTCATTCGGCGGTcccaaat
attgtataaCTGCCCTTAATACATACGTTATACCACTTTTGCACCATATACTTACCACTCCATTTATATACACTTATGTC
AATATTACAGAAAAATCCCCACAAAAATCacctaaacataaaaatattctacttttcaacaataataCATAAACATATTG
GCTTGTGGTAGCAACACTATCATGGTATCACTAACGTAAAAGTTCCTCAATATTGCAATTTGCTTGAACGGATGCTATTT
CAGAATATTTCGTACTTACACAGGCCATACATTAGAATAATATGTCACATCACTGTCGTAACACTCTTTATTCACCGAGC
AATAATACGGTAGTGGCTCAAACTCATGCGGGTGCTATGATACAATTATATCTTATTTCCATTCCCATATGCTAACCGCA
ATATCCTAAAAGCATAACTGATGCATCTTTAATCTTGTATGTGACACTACTCATACGAAGGGACTATATCTAGTCAAGAC
GATACTGTGATAGGTACGTTATTTAATAGGATCTATAACGAAATgtcaaataattttacgGTAATATAACTTATCAGCGG
CGTATACTAAAACGGACGTTACGATATTGTCTCACTTCATCTTACCACCCTCTATCTTATTGCTGATAGAACACTAACCC
CTCAGCTTTATTTCTAGTTACAGTTACACAAAAAACTATGCCAACCCAGAAATCTTGATATTTTACGTGTCAAAAAATGA
GGGTCTCTAAATGAGAGTTTGGTACCATGACTTGTAACTCGCACTGCCCTGATCTGCAATCTTGTTCTTAGAAGTGACGC
ATATTCTATACGGCCCGACGCGACGCgccaaaaaatgaaaaacgAAGCAGCGactcatttttatttaagGACAAAGGTTG
CGAAGCCGCACATTTCCAATTTCATTGTTGTTTATTGGACATACACTGTTAGCTTTATTACCGTCCACGTTTTTTCTACA
ATAGTGTagaagtttctttcttatgTTCATCGTATTCATAAAATGCTTCACGAACACCGTCATTGATCAAATAGgtctat
aatattaatatacatttatataaTCTACGGTATttatatcatcaaaaaaaagtagtttttttattttattttgttcgtta
attttcaatttctatGGAAACCCGTTCGTAAAATTGGCGTTTGTCTCTAGTTTGCGATAGTGTAGATACCGTCCTTGGAT
AGAGCACTGGAGATGGCTGGCTTTAATCTGCTGGAGTACCATGGAACACCGGTGATCATTCTGGTCACTTGGTCTGGAGC
AATACCGGTCAACATGGTGGTGAAGTCACCGTAGTTGAAAACGGCTTCAGCAACTTCGACTGGGTAGGTTTCAGTTGGGT
GGGCGGCTTGGAACATGTAGTATTGGGCTAAGTGAGCTCTGATATCAGAGACGTAGACACCCAATTCCACCAAGTTGACT
CTTTCGTCAGATTGAGCTAGAGTGGTGGTTGCAGAAGCAGTAGCAGCGATGGCAGCGACACCAGCGGCGATTGAAGTTAA
TTTGACCattgtatttgttttgtttgttaGTGCTGATATAAGCTTAACAGGAAAGGAAAGAATAAAGACATATTCTCAAA
GGCATATAGTTGAAGCAGCTCTATTTATACCCATTCCCTCATGGGTTGTTGCTATTTAAACGATCGCTGACTGGCACCAG
TTCCTCATCAAATATTCTCTATATCTCATCTTTCACACAATCTCATTATCTCTATGGAGATGCTCTTGTTTCTGAACGAA
TCATAAATCTTTCATAGGTTTCGTATGTGGAGTACTGTTTTATGGCGCTTATGTGTATTCGTATGCGCAGAATGTGGGAA
TGCCAATTATAGGGGTGCCGAGGTGCCTTATAAAACCCTTTTCTGTGCCTGTGacatttcctttttcggtcaaaaagaat
atccGAATTTTAGATTTGGACCCTCGTACAGAAGCTTATTGTCTAAGCCTGAATTCAGTCTGCTTTAAACGGCTTCCGCG
GAGGAAATATTTCCATCTCTTGAATTCGTACAACATTAAACGTGTGTTGGGAGTCGTATACTGTTAGGGTCTGTAAACTT
GTGAACTCTCGGCAAATGCCTTGGTGCAATTACGTAATTTTAGCCGCTGAGAAGCGGATGGTAATGAGACAAGTTGATAT
CAAACAGATACATATTTAAAAGAGGGTACCGCTAATTTAGCAGGGCAGTATTATTGTAGTTTGATATGTACGGCTAACTG
AACCTAAGTAGGGATATGAGAGTAAGAACGTTCGGCTACTCTTCTTTCTAAGTGGGATTTTTCTTAATCCTTGGATTCTT
AAAAGGTTATTAAAGTTCCGCACAAAGAACGCTTGGAAATCGCATTCATCAAAGAACAACTCTTCGTTTTCCAAACAATC
TTCCCGAAAAAGTAGCCGTTCATTTCCCTTCCGATTTCATTCCTAGACtgccaaatttttcttgctcATTTATAATGATT
GATAAGAATTGTATTTGTGTCCCATTCTCGTAGATAAAATTCTTGGAtgttaaaaaattattattttcttcataaagAAG
CTTTCAAGATATAAGATACGAAATAGGGGTTGATAATTGCATGACAGTAGCTTTAgatcaaaaaggaaagcaTGGAGGGA
AACAGTAAACAGTGAAAATTCTCTTGAGAACCAAAGTAAACCTTCATTGAAGAGCTTCcttaaaaaatttagaaTCTCCC
ATGTCAACGGGTTTCCATACCTCCCCAGCATCATacatcttttttcaaagaaactTCAAATGCCTCTTTTATGCAAGGGG
CAAAATCCTGAAATGACTTAAACTTAGCAGTttcgtcttttttcaaagagaatggttgaagaagaattgtttTGGACGCT
TATTGACAATCTGTTGCATTGATAAAGTACCTACTATCCCAGACTATATTTGTATACAAGTACAAAATTAGGTTTGTTGA
AACAACTTTCCGATCATTGGTGCCCGTATCTGATGTTTTTTTAGTAATTTCTTTGTAAATACAGGGAGTTGTTTCGAAAG
CTTATGAGAAAAATACATGAATGACAGGTAAAAATATTGGCTCGAAAAAGAGGacaaaaagagaaatcaTAAATGAGTAA
ACCCACTTGCTGGACATTATCCAGTAAAGGCTTGGTAGTAACCATAATATTACCCAGGTACGAAACGCTAAGAACTTGAA
AGACTCATAAAACTTCCAGGTTAAgctatttttgaaaatattctgaGGTAAAAGCCATTAAGGTCCAGATAACCAAGGGA
...
Click here to preview GFF file sample_genome.gff 
##gff-version 3
#!gff-spec-version 1.21
#!processor NCBI annotwriter
#!genome-build R64
#!genome-build-accession NCBI_Assembly:GCF_000146045.2
#!annotation-source SGD R64-3-1
##sequence-region Chr01 1 230218
##species https://www.ncbi.nlm.nih.gov/Taxonomy/Browser/wwwtax.cgi?id=559292
Chr01	RefSeq	region	1	230218	.	+	.	ID=Chr01:1..230218;Dbxref=taxon:559292;Name=I;chromosome=I;gbkey=Src;genome=chromosome;mol_type=genomic DNA;strain=S288C
Chr01	RefSeq	telomere	1	801	.	-	.	ID=id-Chr01:1..801;Dbxref=SGD:S000028862;Note=TEL01L%3B Telomeric region on the left arm of Chromosome I%3B composed of an X element core sequence%2C X element combinatorial repeats%2C and a short terminal stretch of telomeric repeats;gbkey=telomere
Chr01	RefSeq	origin_of_replication	707	776	.	+	.	ID=id-Chr01:707..776;Dbxref=SGD:S000121252;Note=ARS102%3B Autonomously Replicating Sequence;gbkey=rep_origin
Chr01	RefSeq	gene	1807	2169	.	-	.	ID=gene-YAL068C;Dbxref=GeneID:851229;Name=PAU8;end_range=2169,.;gbkey=Gene;gene=PAU8;gene_biotype=protein_coding;locus_tag=YAL068C;partial=true;start_range=.,1807
Chr01	RefSeq	mRNA	1807	2169	.	-	.	ID=rna-NM_001180043.1;Parent=gene-YAL068C;Dbxref=GeneID:851229,Genbank:NM_001180043.1;Name=NM_001180043.1;end_range=2169,.;gbkey=mRNA;gene=PAU8;locus_tag=YAL068C;partial=true;product=seripauperin PAU8;start_range=.,1807;transcript_id=NM_001180043.1
Chr01	RefSeq	exon	1807	2169	.	-	.	ID=exon-NM_001180043.1-1;Parent=rna-NM_001180043.1;Dbxref=GeneID:851229,Genbank:NM_001180043.1;end_range=2169,.;gbkey=mRNA;gene=PAU8;locus_tag=YAL068C;partial=true;product=seripauperin PAU8;start_range=.,1807;transcript_id=NM_001180043.1
Chr01	RefSeq	CDS	1807	2169	.	-	0	ID=cds-NP_009332.1;Parent=rna-NM_001180043.1;Dbxref=SGD:S000002142,GeneID:851229,Genbank:NP_009332.1;Name=NP_009332.1;Note=hypothetical protein%3B member of the seripauperin multigene family encoded mainly in subtelomeric regions;experiment=EXISTENCE:mutant phenotype:GO:0030437 ascospore formation [PMID:12586695],EXISTENCE:mutant phenotype:GO:0045944 positive regulation of transcription by RNA polymerase II [PMID:12586695];gbkey=CDS;gene=PAU8;locus_tag=YAL068C;product=seripauperin PAU8;protein_id=NP_009332.1
Chr01	RefSeq	gene	2480	2707	.	+	.	ID=gene-YAL067W-A;Dbxref=GeneID:1466426;Name=YAL067W-A;end_range=2707,.;gbkey=Gene;gene_biotype=protein_coding;locus_tag=YAL067W-A;partial=true;start_range=.,2480
Chr01	RefSeq	mRNA	2480	2707	.	+	.	ID=rna-NM_001184582.1;Parent=gene-YAL067W-A;Dbxref=GeneID:1466426,Genbank:NM_001184582.1;Name=NM_001184582.1;end_range=2707,.;gbkey=mRNA;locus_tag=YAL067W-A;partial=true;product=uncharacterized protein;start_range=.,2480;transcript_id=NM_001184582.1
Chr01	RefSeq	exon	2480	2707	.	+	.	ID=exon-NM_001184582.1-1;Parent=rna-NM_001184582.1;Dbxref=GeneID:1466426,Genbank:NM_001184582.1;end_range=2707,.;gbkey=mRNA;locus_tag=YAL067W-A;partial=true;product=uncharacterized protein;start_range=.,2480;transcript_id=NM_001184582.1
Chr01	RefSeq	CDS	2480	2707	.	+	0	ID=cds-NP_878038.1;Parent=rna-NM_001184582.1;Dbxref=SGD:S000028593,GeneID:1466426,Genbank:NP_878038.1;Name=NP_878038.1;Note=hypothetical protein%3B identified by gene-trapping%2C microarray-based expression analysis%2C and genome-wide homology searching;gbkey=CDS;locus_tag=YAL067W-A;product=uncharacterized protein;protein_id=NP_878038.1
Chr01	RefSeq	gene	7235	9016	.	-	.	ID=gene-YAL067C;Dbxref=GeneID:851230;Name=SEO1;end_range=9016,.;gbkey=Gene;gene=SEO1;gene_biotype=protein_coding;locus_tag=YAL067C;partial=true;start_range=.,7235
...

Tutorial

  1. Navigate to the project folder.

    $ cd CROPSR

  2. Run CROPSR with the sample data (code is available below).

    $ python3 CROPSR.py -f sample_data/sample_genome.fa -g sample_data/sample_genome.gff -o sample_data/sample_genome_output.csv --cas9 -v

    Note that the verbose flag was left on. This will cause the terminal to print notifications during the process, however, it means you will not be able to utilize the terminal window until it is finished. If you close the terminal window while the process is running, it will cause an interruption.

    Click here to learn how to run this process in the background 
    $ python3 CROPSR.py -f sample_data/sample_genome.fa -g sample_data/sample_genome.gff -o sample_data/sample_genome_output.csv --cas9 &

    In this variation, the verbose flag was removed, and a & was added at the end of the command. This will free your terminal window to perform other tasks or be closed. This is the recommended approach when running real data, as the process may take more than a day to finish. Make sure the computer remains powered on for the entirety of the process.

  3. While CROPSR is running (with the verbose flag active), you should see the following appear in your terminal:

    ################################################################################
##                                                                            ##
##                                                                            ##
##          .o88b.   d8888b.    .d88b.    d8888b.   .d8888.   d8888b.         ##
##         d8P  Y8   88  `8D   .8P  Y8.   88  `8D   88'  YP   88  `8D         ##
##         8P        88oobY'   88    88   88oodD'   `8bo.     88oobY'         ##
##         8b        88`8b     88    88   88ººº       `Y8b.   88`8b           ##
##         Y8b  d8   88 `88.   `8b  d8'   88        db   8D   88 `88.         ##
##          `Y88P'   88   YD    `Y88P'    88        `8888Y'   88   YD         ##
##                                                                            ##
##                                                                            ##
################################################################################
U.S. Dept. of Energy's Center for Advanced Bioenergy and Bioproducts Innovation
University of Illinois at Urbana-Champaign

        You are currently utilizing the following settings:

        CROPSR version:                                 1.11b
        Path to genome file in FASTA format:            sample_data/sample_genome.fa
        Path to output file:                            sample_data/output.csv
        Length of the gRNA sequence:                    20
        Length of flanking region for verification:     200
        Number of available CPUs:                       12
        Path to annotation file in GFF format:          /sample_data/sample_genome.gff3
        Path to annotation_info file in TXT format:     None
        Designing for CRISPR system:
            Streptococcus pyogenes Cas9                 True
        
Genome file sample_data/sample_genome.fa successfully imported
formatting genome
Genome file sample_data/sample_genome.fa successfully formatted
The genome was successfully converted to a dictionary
Annotation file sample_data/sample_genome.gff successfully imported
Annotation database successfully generated

        Initiating PAM site detection.
        
        Please wait, this may take a while...
        

            17314 Cas9 PAM sites were found on Chr01
            
The output file has been generated at sample_data/output.csv

  4. After the process is complete, you should have access to the generated output file in CSV format.

    The folder structure should be similar to what is represented below:

    +-- README.md
+-- LICENCE.md
+-- CROPSR.py
+-- cropsr_functions.py
+-- prmrdsgn2.py
+-- sample_data/
    +-- sample_genome.fa
    +-- sample_genome.gff
    +-- sample_genome_output.csv
+-- .DS_Store
+-- .gitignore
    Click here for a preview of sample_genome_output.csv 
    crispr_id,crispr_sys,sequence,long_sequence,chromosome,start_pos,end_pos,cutsite,strand,on_site_score,features
A01NW7FGPN,cas9,GGUUAGAUUAGGGCUGUGUU,GCCAGGGUUAGAUUAGGGCUGUGUUAGGGU,Chr01,77,97,94,+,0.0388536288320188,,completed
A01QLYYDXZ,cas9,GUGCGUACGUAAAAUCAGUA,UCCGUGUGCGUACGUAAAAUCAGUAUACAA,Chr01,411,431,428,+,0.5402860690510768,,completed
A01RVDM36X,cas9,GGAGUGAAGUGGAAUCUGAG,GCCAUGGAGUGAAGUGGAAUCUGAGAGUAG,Chr01,471,491,488,+,0.6091206773441749,,completed
A011BCL3O8,cas9,GCAUAAUGAUGUGAGUGCAU,GCCGUGCAUAAUGAUGUGAGUGCAUUUGGU,Chr01,521,541,538,+,0.05499563386100348,,completed
A01J39N6WJ,cas9,UGAGGCAAGUGCCGUGCAUA,ACCGCUGAGGCAAGUGCCGUGCAUAAUGAU,Chr01,536,556,553,+,0.1199299810564199,,completed
A01AG1OHUM,cas9,AUGAGAUAUAGAUAUCAAAA,GCCGAAUGAGAUAUAGAUAUCAAAAUGUGG,Chr01,605,625,622,+,0.7178898545540503,,completed
A01HBGMKT6,cas9,CGAAUGAGAUAUAGAUAUCA,ACCGCCGAAUGAGAUAUAGAUAUCAAAAUG,Chr01,608,628,625,+,0.16003821718975292,,completed
A01LPJZIOI,cas9,UAUGUUUAUGataataacaa,GCCAAUAUGUUUAUGataataacaactttt,Chr01,777,797,794,+,0.8846303027545066,,completed
A019RQ1MNC,cas9,AAGCCAAUAUGUUUAUGata,ACCACAAGCCAAUAUGUUUAUGataataac,Chr01,784,804,801,+,0.2709374516526445,,completed
...

Back to top

Disclosures

This work was funded by the DOE Center for Advanced Bioenergy and Bioproducts Innovation (U.S. Department of Energy, Office of Science, Office of Biological and Environmental Research under Award Number DE-SC0018420). Any opinions, findings, and conclusions or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the views of the U.S. Department of Energy.

Back to top