# This CITATION.cff file was generated with cffinit.
# Visit https://bit.ly/cffinit to generate yours today!

cff-version: 1.2.0
title: AllInOne
message: Make Your Couch a Data Pre-Processing Center
type: software
authors:
  - given-names: Mohsen
    family-names: Yoosefzadeh Najafabadi
    email: myoosefz@uoguelph.ca
    affiliation: Department of Plant Agriculture, University of Guelph, Guelph, ON N1G 2W1, Canada
    orcid: 'https://orcid.org/0000-0003-3631-4493'
  - orcid: 'https://orcid.org/0000-0003-2087-4306'
    email: Alihdr@stu.usc.ac.ir
    family-names: Heidari
    given-names: Ali
  - given-names: Istvan
    family-names: Rajcan
    email: Irajcan@uoguelph.ca
    orcid: 'https://orcid.org/0000-0001-5156-2482'
    affiliation: Department of Plant Agriculture, University of Guelph, Guelph, ON N1G 2W1, Canada
identifiers:
  - type: url
    value: 'https://github.com/MohsenYN/AllInOne'
repository-code: 'https://github.com/MohsenYN/AllInOne'
url: 'https://allinone.shinyapps.io/allinone/'
abstract: >-
  Plant breeding is a great mixture of art and science that
  has been around for centuries and plays an important role
  in solving many of the world’s agricultural problems. The
  ultimate goal for most plant breeding programs is to
  develop new varieties that are better adapted for certain
  environments, more resistant to biotic/abiotic stresses,
  and/or have increased yields. The process usually starts
  with selecting and crossing two varieties as parental
  lines with specific features to create a population with a
  high recombination rate. Afterward, the created population
  is evaluated for several generations to select the
  superior varieties based on the plant breeding goal. The
  first evaluation of the breeding population mainly relies
  on the phenotypic data derived from different plant
  varieties growing in multi-locations for several years.
  The collected phenotypic data can be further used in
  combination with the environment and omics data to
  increase the accuracy of the breeding decisions.
  Therefore, the accuracy, quality, and nature of phenotypic
  data are the three most important factors in capturing
  true signals and avoiding false interpretations. In
  addition, the proper selection of statistical analysis
  methods is of high paramount in making the right decision
  to select superior genotypes based on a trait of
  interest. 

  The accuracy of phenotypic data highly depends on how
  plant breeders collect data. Conventional data collection
  methods can bring more noise to the phenotypic data than
  high throughput methods. In addition, the contribution of
  more data collectors in measuring phenotypic data may
  decrease the level of accuracy. Moreover, the existence
  and abundance of genotypes that exhibit a particular
  phenotypic characteristic is an important determinant of
  the extent to which analysis methods can be used to select
  superior genotypes. Therefore, as plant breeders have been
  dealing with a large number of genotypes in different
  locations over several years, it is necessary to
  pre-process the phenotypic datasets before selecting any
  methods or making any decision. Pre-processing is one of
  the important procedures for increasing the quality and
  accuracy of the field phenotypic data, which is usually
  done in several steps, such as 1) detecting missing
  patterns in the dataset, 2) imputing missing data using
  different statistical methods, 3) data visualization in
  order to check data patterns and distribution, 4)
  detecting and refining outliers, 5) estimating
  correlations between dependant variables and also with and
  within independent variables, 6) normalizing data based on
  the optimum normalization methods for a given dataset, 6)
  estimating heritability and conducting spatial analysis,
  and finally, 7) calculating best linear unbiased
  prediction (BLUP) or/and best linear unbiased estimator
  (BLUE) based on the goal of the plant breeding program. 

  Several packages in different languages (mainly in R) have
  been created to handle different pre-processing steps,
  such as 1) MICE to deal with missing data, 2)
  Bestormalizer to normalize the datasets using different
  methods, 3) lme4 and ASReml R to handle variance analysis
  using different experimental designs, 4) ggplot for data
  visualization, etc. However, none of them are able to
  handle at least most of the pre-processing steps in plant
  breeding phenotypic datasets. In addition, they all
  require medium to advanced coding knowledge to run and
  adjust the functions based on the breeding preference.
  Furthermore, more sophisticated packages such as ASReml R
  are not free and require an annual renewal fee. Moreover,
  none of them provided a dynamical graphical interference
  for creating plots, detecting and refining outliers in a
  live mode, and using different datasets without directly
  importing from a file. 

  Here, we introduce the AllInOne package as an open-Source,
  breeder-friendly, analytical R package for pre-processing
  phenotypic data. The basis of AllInOne is to utilize
  different R packages and develop the pipeline for
  pre-processing the phenotypic datasets in an accurate,
  easy, and timely manner without any coding skills
  required. In addition, several new features and abilities
  were added in AllInOne that can complement previously
  developed packages in this area.
keywords:
  - Pre-processing
  - plant breeding
  - data collection
  - curation
  - R shiny package
  - agriculture
license: MIT
version: 1.8
date-released: '2023-1-18'