# ATG
_ATG_ is a library and standalone CLI tool to handle and convert different data formats used in Genomics and Transcriptomics. The library provides convenient APIs to parse GTF and RefGene data and work with the resulting transcripts for all kind of downstream analyses.
The binary can be used to convert between GTF and RefGene data, generate bed files for transcripts or generate the nucleotide sequence as Fasta.
The main purpose is actually just that - convert between GTF and RefGene file formats. Surprsingly, there are not many tools to do this properly. Even _atg_ does not handle all edge cases of GTF - but I tried to handle as many as possible. In addition, transcripts can also be written in bed format or as fasta.
The project started only because I wanted to learn Rust. You will see that some sections have really bad code, others will have some better and more improved code. Overall, I'm still very new to Rust and I'm sure I fell for many traps and use lots of unidiomatic code. I'm happy for any feedback and improvement suggestions.
The library is still in its infancy but works so far and can handle what it's supposed to do. The current API is probably going to change a lot in future updates, so be careful of using _atg_ in production or other critical workflows.
## Usage
### ATG command line tool
#### Install
##### cargo
The easiest way to install _ATG_ is to use `cargo` (if you have `cargo` and `rust` installed)
```bash
cargo install atg
```
##### Pre-built binaries
You can download pre-built binaries for Linux and Mac (M1) from [Github](https://github.com/anergictcell/atg/releases). Save the downloaded binaries in a folder in your `PATH` or create a symlink as `atg`.
##### From source
You can also build _ATG_ from source (if you have the rust toolchains installed):
```bash
git clone https://github.com/anergictcell/atg.git
cd atg
cargo build --release
````
#### Run
Convert a GTF file to a RefGene file
```bash
atg --from gtf --to refgene --input /path/to/input.gtf --output /path/to/output.refgene
```
Convert RefGene to GTF
```bash
atg --from refgene --to gtf --input /path/to/input.refgene --output /path/to/output.gtf
```
Convert RefGene to bed
```bash
atg --from refgene --to bed --input /path/to/input.refgene --output /path/to/output.bed
```
##### Supported `--output` formats
###### gtf
Output in GTF format.
```text
chr9 ncbiRefSeq.2021-05-17 transcript 74526555 74600974 . + . gene_id "C9orf85"; transcript_id "NM_001365057.2"; gene_name "C9orf85";
chr9 ncbiRefSeq.2021-05-17 exon 74526555 74526752 . + . gene_id "C9orf85"; transcript_id "NM_001365057.2"; exon_number "1"; exon_id "NM_001365057.2.1"; gene_name "C9orf85";
chr9 ncbiRefSeq.2021-05-17 5UTR 74526555 74526650 . + . gene_id "C9orf85"; transcript_id "NM_001365057.2"; exon_number "1"; exon_id "NM_001365057.2.1"; gene_name "C9orf85";
chr9 ncbiRefSeq.2021-05-17 CDS 74526651 74526752 . + 0 gene_id "C9orf85"; transcript_id "NM_001365057.2"; exon_number "1"; exon_id "NM_001365057.2.1"; gene_name "C9orf85";
chr9 ncbiRefSeq.2021-05-17 exon 74561922 74562028 . + . gene_id "C9orf85"; transcript_id "NM_001365057.2"; exon_number "2"; exon_id "NM_001365057.2.2"; gene_name "C9orf85";
chr9 ncbiRefSeq.2021-05-17 CDS 74561922 74562026 . + 0 gene_id "C9orf85"; transcript_id "NM_001365057.2"; exon_number "2"; exon_id "NM_001365057.2.2"; gene_name "C9orf85";
...
```
You can specify the value of the `source` column manually using the `--gtf-source`/`-g` option. Defaults to `atg`
###### refgene
Output in the refGene format, as used by some UCSC and NCBI RefSeq services
```text
0 NM_001101.5 chr7 - 5566778 5570232 5567378 5569288 6 5566778,5567634,5567911,5568791,5569165,5570154, 5567522,5567816,5568350,5569031,5569294,5570232, 0 ACTB cmpl cmpl 0,1,0,0,0,-1,
0 NM_001203247.2 chr7 - 148504474 148581383 148504737 148544390 20 148504474,148506162,148506401,148507424,148508716,148511050,148512005,148512597,148513775,148514313,148514968,148516687,148523560,148524255,148525831,148526819,148529725,148543561,148544273,148581255, 148504798,148506247,148506482,148507506,148508812,148511229,148512131,148512638,148513870,148514483,148515209,148516779,148523724,148524358,148525972,148526940,148529842,148543690,148544397,148581383, 0 EZH2 cmpl cmpl 2,1,1,0,0,1,1,2,0,1,0,1,2,1,1,0,0,0,0,-1,
0 NM_001203248.2 chr7 - 148504474 148581383 148504737 148544390 20 148504474,148506162,148506401,148507424,148508716,148511050,148512005,148512597,148513775,148514313,148514968,148516687,148523560,148524255,148525831,148526819,148529725,148543588,148544273,148581255, 148504798,148506247,148506482,148507506,148508812,148511229,148512131,148512638,148513870,148514483,148515209,148516779,148523724,148524358,148525972,148526940,148529842,148543690,148544397,148581383, 0 EZH2 cmpl cmpl 2,1,1,0,0,1,1,2,0,1,0,1,2,1,1,0,0,0,0,-1,
0 NM_001354750.2 chr11 + 113930432 114127487 113934022 114121277 7 113930432,113933932,114027058,114057673,114112888,114117919,114121047, 113930864,113935290,114027156,114057760,114113059,114118087,114127487, 0 ZBTB16 cmpl cmpl -1,0,2,1,1,1,1,
```
###### fasta
Writes the cDNA sequence of all transcripts into one file. Please note that the sequence is stranded.
This target format requires a reference genome fasta file that must be specified using `--reference`/`-r`.
*This output allows different `--fasta-format` options:*
- `transcript`: The full transcript sequence (from the genomic start to end position, including introns)
- `exons`: The cDNA sequence of the processed transcript, i.e. the sequence of all exons, including non-coding exons.
- `cds` (default): The CDS of the transcript
```text
>NM_007298.3 BRCA1
ATGGATTTATCTGCTCTTCGCGTTGAAGAAGTACAAAATGTCATTAATGC
TATGCAGAAAATCTTAGAGTGTCCCATCTGTCTGGAGTTGATCAAGGAAC
CTGTCTCCACAAAGTGTGACCACATATTTTGCAAATTTTGCATGCTGAAA
CTTCTCAACCAGAAGAAAGGGCCTTCACAGTGTCCTTTATGTAAGAATGA
TATAACCAAAAGGAGCCTACAAGAAAGTACGAGATTTAGTCAACTTGTTG
...
>NM_001365057.2 C9orf85
ATGAGCTCCCAGAAAGGCAACGTGGCTCGTTCCAGACCTCAGAAGCACCA
GAATACGTTTAGCTTCAAAAATGACAAGTTCGATAAAAGTGTGCAGACCA
AGAAAATTAATGCAAAACTTCATGATGGAGTATGTCAGCGCTGTAAAGAA
GTTCTTGAGTGGCGTGTAAAATACAGCAAATACAAACCATTATCAAAACC
TAAAAAGTGA
...
```
###### fasta-split
Like `fasta` above, but one file for each transcript. Instead of an output file, you must specify an output directory, _ATG_ will save each transcript as `<Transcript_name>.fasta`, e.g.: `NM_001365057.2.fasta`.
This target format requires a reference genome fasta file that must be specified using `--reference`/`-r`.
*This output allows different `--fasta-format` options:*
- `transcript`: The full transcript sequence (from the genomic start to end position, including introns)
- `exons`: The cDNA sequence of the processed transcript, i.e. the sequence of all exons, including non-coding exons.
- `cds` (default): The CDS of the transcript
###### feature-sequence
cDNA sequence of each feature (5' UTR, CDS, 3'UTR), each in a separate row.
This target format requires a reference genome fasta file that must be specified using `--reference`/`-r`.
```text
BRCA1 NM_007298.3 chr17 41196311 41197694 - 3UTR CTGCAGCCAGCCAC...
BRCA1 NM_007298.3 chr17 41197694 41197819 - CDS CAATTGGGCAGATGTGTG...
BRCA1 NM_007298.3 chr17 41199659 41199720 - CDS GGTGTCCACCCAATTGTG...
BRCA1 NM_007298.3 chr17 41201137 41201211 - CDS ATCAACTGGAATGGATGG...
BRCA1 NM_007298.3 chr17 41203079 41203134 - CDS ATCTTCAGGGGGCTAGAA...
BRCA1 NM_007298.3 chr17 41209068 41209152 - CDS CATGATTTTGAAGTCAGA...
BRCA1 NM_007298.3 chr17 41215349 41215390 - CDS GGGTGACCCAGTCTATTA...
BRCA1 NM_007298.3 chr17 41215890 41215968 - CDS ATGCTGAGTTTGTGTGTG...
BRCA1 NM_007298.3 chr17 41219624 41219712 - CDS ATGCTCGTGTACAAGTTT...
BRCA1 NM_007298.3 chr17 41222944 41223255 - CDS AGGGAACCCCTTACCTGG...
C9orf85 NM_001365057.2 chr9 74526555 74526650 + 5UTR ATTGACAGAA...
C9orf85 NM_001365057.2 chr9 74526651 74526752 + CDS ATGAGCTCCCAGAA...
C9orf85 NM_001365057.2 chr9 74561922 74562028 + CDS AAAATTAATGCAAA...
C9orf85 NM_001365057.2 chr9 74597573 74597573 + CDS A
C9orf85 NM_001365057.2 chr9 74597574 74600974 + 3UTR TGGAGTCTCC...
```
###### raw
This is mainly useful for debugging, as it gives a quick glimpse into the Exons and CDS coordinates of the transcripts.
###### bin
Save Transcripts in _ATG_ binary format for faster re-reading.
#### Tips
Reading in GTF files is rather slow, due to the complexity of the format. If you need to repeatedly read in data from GTF, I recommend to generate a RefGene or binary file once and use this as input for subsequent steps.
You can change the verbosity, by adding `-v` (show info messages), `-vv` (debug), `-vvv` (trace)
On most Linux systems, you can use `--input /dev/stdin` and/or `--output /dev/stdout` to pipe into and out of atg.
Of course, all commands also have shorthand parameters:
- `-f`, `--from`
- `-t`, `--to`
- `-i`, `--input`
- `-o`, `--output`
### ATG as library
[The library API is mostly documented inline and available on docs.rs](https://docs.rs/atg)
#### Examples
##### Convert GTF to RefGene
```no_run
use atg::gtf::Reader;
use atg::refgene::Writer;
use atg::models::{TranscriptRead, TranscriptWrite};
let mut reader = Reader::from_file("path/to/input.gtf")
.unwrap_or_else(|_| panic!("Error opening input file."));
let mut writer = Writer::from_file("path/to/output.refgene")
.unwrap_or_else(|_| panic!("Unable to open output file"));
let transcripts = reader.transcripts()
.unwrap_or_else(|err| panic!("Error parsing GTF: {}", err));
match writer.write_transcripts(&transcripts) {
Ok(_) => println!("Success"),
Err(err) => panic!("Error writing RefGene file: {}", err)
};
```
## ToDo / Next tasks
- [x] Add to crates.io
- [x] Bed module to generate bed files with exons and introns
- [ ] Compare transcripts from two different inputs
- [x] Add fasta reading for nt and aa sequence outputs
- [x] Binary data format
## Known issues
### GTF parsing
- [ ] NM_001371720.1 has two book-ended exons (155160639-155161619 || 155161620-155162101). During input parsing, book-ended features are merged into one exon