---
title: "Converting Common Data Formats to Phyloseq and TreeSummarizedExperiment"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Data Import}
  %\VignetteEncoding{UTF-8}
  %\VignetteEngine{knitr::rmarkdown}
editor_options: 
  chunk_output_type: console
---

<style>
body {text-align: justify}
</style>

```{r, include = FALSE}
knitr::opts_chunk$set(
  message = FALSE,
  digits = 3,
  collapse = TRUE,
  comment = "#>"
)
options(digits = 3)
```

The "dar" package is a versatile and user-friendly tool designed to accept 
inputs in a variety of formats. It primarily utilizes the `phyloseq` format but 
also supports the `TreeSummarizedExperiment` format. This flexibility allows 
users to conduct differential abundance analysis smoothly, irrespective of 
their initial data format. To facilitate this, a detailed guide is available 
to aid users in converting other prevalent data formats, such as `biome`, 
`mothur`, `metaphlan`, and more, into the necessary `phyloseq` or 
`TreeSummarizedExperiment` formats. 

```{r}
suppressPackageStartupMessages(library(mia))
suppressPackageStartupMessages(library(phyloseq))
```

## Importing Data from `biome` Format

The `biome` format is a commonly used format in bioinformatics to represent 
microbiome sequencing data. Here's how you can import data in `biome` format to 
both `phyloseq` and `TreeSummarizedExperiment.`


### To Phyloseq

To convert data from the `biome` format to the `phyloseq` format, you can use 
the `phyloseq::import_biom()` function. Here's a step-by-step example of how to 
perform this conversion:

```{r}
# Example of a rich dense biom file
rich_dense_biom  <-
  system.file("extdata", "rich_dense_otu_table.biom",  package = "phyloseq")

# Import biom as a phyloseq-class object
phy <- phyloseq::import_biom(
  rich_dense_biom,  
  parseFunction = parse_taxonomy_greengenes
)

phy

# Print sample_data
phyloseq::sample_data(phy)

# Print tax_table
phyloseq::tax_table(phy)

# Recipe init
rec <- dar::recipe(phy, var_info = "BODY_SITE", tax_info = "Genus")

rec
```


### To TreeSummarizedExperiment

To convert data from the `biome` format to the `TreeSummarizedExperiment` format, 
you can use the `mia::importBIOM()` function. Here's a step-by-step example 
of how to perform this conversion:

```{r}
# Example of a rich dense biom file
rich_dense_biom  <-
  system.file("extdata", "rich_dense_otu_table.biom",  package = "phyloseq")

# Import biom as a phyloseq-class object
tse <- mia::importBIOM(rich_dense_biom)

tse

# Print sample_data
colData(tse)

# Print tax_table
rowData(tse)

# Change the column names of the tax_table
colnames(rowData(tse)) <- 
  c("Kingdom", "Phylum", "Class", "Order", "Family", "Genus", "Species")

rowData(tse)

# Recipe init
rec <- dar::recipe(tse, var_info = "BODY_SITE", tax_info = "Genus")

rec
```


## Importing Data from `qiime` Format

The `qiime` format is another commonly used format in bioinformatics for 
microbiome sequencing data. Here's how you can import data in `qiime` format to
both `Phyloseq` and `TreeSummarizedExperiment.`


### To Phyloseq

To convert data from the `qiime` format to the `Phyloseq` format, you can use 
the `phyloseq::import_qiime()` function. Here's a step-by-step example of how 
to perform this conversion:

```{r}
# Import QIIME data
phy_qiime <- phyloseq::import_qiime(
  otufilename = system.file("extdata", "GP_otu_table_rand_short.txt.gz", package = "phyloseq"),
  mapfilename = system.file("extdata", "master_map.txt", package = "phyloseq"),
  treefilename = system.file("extdata", "GP_tree_rand_short.newick.gz", package = "phyloseq")
)

phy_qiime

# Recipe init
rec <- dar::recipe(phy_qiime, var_info = "SampleType", tax_info = "Genus")

rec
```


### To TreeSummarizedExperiment

To convert data from the `qiime` format to the `TreeSummarizedExperiment` format, 
you can use the `mia::importQIIME2()` function. Here's a step-by-step example 
of how to perform this conversion:

```{r}
# Import QIIME data to tse
tse_qiime <- mia::importQIIME2(
  featureTableFile = system.file("extdata", "table.qza", package = "mia"),
  taxonomyTableFile = system.file("extdata", "taxonomy.qza", package = "mia"),
  sampleMetaFile = system.file("extdata", "sample-metadata.tsv", package = "mia"),
  refSeqFile = system.file("extdata", "refseq.qza", package = "mia"),
  phyTreeFile = system.file("extdata", "tree.qza", package = "mia")
)

tse_qiime

# Recipe init
rec <- dar::recipe(tse_qiime, var_info = "body.site", tax_info = "genus")

rec
```


## Importing Data from `mothur` Format

The `mothur` format is another commonly used format in bioinformatics for 
microbiome sequencing data. Here's how you can import data in `mothur` format 
to both `Phyloseq` and `TreeSummarizedExperiment.`


### To Phyloseq

To convert data from the `mothur` format to the `Phyloseq` format, you can use 
the `phyloseq::import_mothur()` function. Here's a step-by-step example of how 
to perform this conversion:

```{r}
# Import Mothur data
phy_mothur <- phyloseq::import_mothur(
  mothur_list_file = system.file("extdata", "esophagus.fn.list.gz", package = "phyloseq"),
  mothur_group_file = system.file("extdata", "esophagus.good.groups.gz", package = "phyloseq"),
  mothur_tree_file = system.file("extdata", "esophagus.tree.gz", package = "phyloseq")
)

phy_mothur

# Recipe init
rec <- dar::recipe(phy_mothur)

rec
```


### To TreeSummarizedExperiment

To convert data from the `mothur` format to the `TreeSummarizedExperiment` 
format, you can use the `mia::importMothur()` function. Here's a step-by-step 
example of how to perform this conversion:

```{r}
# Import Mothur data to TreeSummarizedExperiment
tse_mothur <- mia::importMothur(
  sharedFile = system.file("extdata", "mothur_example.shared", package = "mia"),
  taxonomyFile = system.file("extdata", "mothur_example.cons.taxonomy", package = "mia"),
  designFile = system.file("extdata", "mothur_example.design", package = "mia")
) |> methods::as("TreeSummarizedExperiment")

tse_mothur

# Recipe init
rec <- dar::recipe(tse_mothur, var_info = "drug", tax_info = "Genus")

rec
```


## Importing Data from `metaphlan` Format

The `metaphlan` format is another commonly used format in bioinformatics for 
microbiome sequencing data. Here's how you can import data in `metaphlan` 
format to `TreeSummarizedExperiment.`

### To TreeSummarizedExperiment

To convert data from the `metaphlan` format to the `TreeSummarizedExperiment` 
format, you can use the `mia::importMetaPhlAn()` function. Here's a 
step-by-step example of how to perform this conversion:

```{r}
# Importing data from Metaphlan
tse_metaphlan <- mia::importMetaPhlAn(
  file = system.file("extdata", "merged_abundance_table.txt", package = "mia")
)

# Recipe init
tse_metaphlan <- TreeSummarizedExperiment::TreeSummarizedExperiment(
  assays = list(counts = SummarizedExperiment::assay(tse_metaphlan)),
  rowData = SummarizedExperiment::rowData(tse_metaphlan),
  colData = SummarizedExperiment::colData(tse_metaphlan) |> 
    as.data.frame() |> 
    dplyr::mutate(condition = rep(c("A", "B"), times = 3))
)

rec <- dar::recipe(tse_metaphlan, var_info = "condition", tax_info = "Genus")

rec
```

## Conclusion

In this guide, we have explored various methods for importing microbiome 
sequencing data from different formats into `Phyloseq` and 
`TreeSummarizedExperiment.` We've covered the `biome`, `qiime`, `mothur`, 
`metaphlan`, and `humann` formats, providing step-by-step examples for each.

The flexibility of these tools allows for a smooth transition between different
data formats, making it easier to conduct your analysis irrespective of the
initial data format. By following the steps outlined in this guide, you should 
be able to successfully convert your data and carry out further differential 
abundance analysis.

Remember, the specific details of your data may require you to adjust the 
parameters in the import functions. Always inspect your data after conversion 
to ensure it has been imported correctly.


## Session info

```{r}
devtools::session_info()
```