kfukuta/wf-transcriptomes-v202
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge branch 'jaffal_docs' into 'dev'

Browse Source 
Update docs to remove JAFFAL install instructions

See merge request epi2melabs/workflows/wf-transcriptomes!89
This commit is contained in:
Neil Horner 2023-02-24 16:49:28 +00:00
commit 5c0a93bc17
3 changed files with 32 additions and 67 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

47
README.md
View File

@ -172,48 +172,33 @@ __Note__: edlib is set by default in the config as it's quite a lot faster. Howe
### Fusion detection
JAFFAL from the [JAFFA](https://github.com/Oshlack/JAFFA)
package is used to identify potential fusion transcripts. To get this this working, there are a couple of things that need doing first.
package is used to identify potential fusion transcripts.
**Install JAFFA**
to install JAFFA and it's dependencies run the folllowing:
```shell
cd wf-transcriptomes/
./subworkflows/JAFFAL/install_jaffa.sh
```
**Prepare JAFFAL reference data**
To use pre-processed reference files for the hg38 genome and GENCODE v22 annotation (as used in the JFFAAL paper),
In order to use JAFFAL, reference files must first be downloaded.
To use pre-processed hg38 genome and GENCODE v22 annotation files (as used in the JAFFAL paper)
do:
```shell
mkdir jaffal_data_dir
cd jaffal_data_dir/
wf-transcriptomes/subworkflows/JAFFAL/load_jaffal_references.sh
sh path/to/wf-transcriptomes/subworkflows/JAFFAL/download_jaffal_references.sh
````
To use alternative genome and annotation files, they should be prepared as described
[here](https://github.com/Oshlack/JAFFA/wiki/FAQandTroubleshooting#how-can-i-generate-the-reference-files-for-a-non-supported-genome)
**Specifying the location of the JAFFA code and reference directories**
`--jaffal_dir`
Full path to the directory made by running install_jaffa.sh as shown above. eg: /home/wf-trnascriptomes/JAFFA
`--jaffal_refBase`
The directory containing the reference data prepared for use with JAFFAL
Then the path to the directory containing the downloaded reference data must be specified with
`--jaffal_refBase`.
**JAFFAL annotation and genome files**
**Using alternative genome and annotation files**
The prepared JAFFAL reference files will look something like `hg38_chr20_genCode22.fa`. To enable JAFFAL to find these
files `--jaffal_genome` should be set to `hg38_chr20` and `--jaffal_annotation` to `genCode22`
These should be prepared as described
[here](https://github.com/Oshlack/JAFFA/wiki/FAQandTroubleshooting#how-can-i-generate-the-reference-files-for-a-non-supported-genome).
The resulting JAFFAL reference files will look something like `hg38_genCode22.fa`. The following options enable JAFFAL to find these
files:
`--jaffal_genome` optional (default: `hg38`)
`--jaffal_annotation` optional (default: `genCode22`)
__JAFFAL Notes__:
g++ must be installed. JAFFAL is not currently working on Mac M1 (osx-arm64 architecture). If there are no fusion transcripts
detected, the workflow will terminate with an error at the JAFFAL stage. If this happens,
skip the JAFFAL stage by omitting ` --jaffal_refBase`
__Note__: JAFFAL is not currently working on Mac M1 (osx-arm64 architecture).
### Differential Expression

47
docs/quickstart.md
View File

@ -82,48 +82,33 @@ __Note__: edlib is set by default in the config as it's quite a lot faster. Howe
### Fusion detection
JAFFAL from the [JAFFA](https://github.com/Oshlack/JAFFA)
package is used to identify potential fusion transcripts. To get this this working, there are a couple of things that need doing first.
package is used to identify potential fusion transcripts.
**Install JAFFA**
to install JAFFA and it's dependencies run the folllowing:
```shell
cd wf-transcriptomes/
./subworkflows/JAFFAL/install_jaffa.sh
```
**Prepare JAFFAL reference data**
To use pre-processed reference files for the hg38 genome and GENCODE v22 annotation (as used in the JFFAAL paper),
In order to use JAFFAL, reference files must first be downloaded.
To use pre-processed hg38 genome and GENCODE v22 annotation files (as used in the JAFFAL paper)
do:
```shell
mkdir jaffal_data_dir
cd jaffal_data_dir/
wf-transcriptomes/subworkflows/JAFFAL/load_jaffal_references.sh
sh path/to/wf-transcriptomes/subworkflows/JAFFAL/download_jaffal_references.sh
````
To use alternative genome and annotation files, they should be prepared as described
[here](https://github.com/Oshlack/JAFFA/wiki/FAQandTroubleshooting#how-can-i-generate-the-reference-files-for-a-non-supported-genome)
**Specifying the location of the JAFFA code and reference directories**
`--jaffal_dir`
Full path to the directory made by running install_jaffa.sh as shown above. eg: /home/wf-trnascriptomes/JAFFA
`--jaffal_refBase`
The directory containing the reference data prepared for use with JAFFAL
Then the path to the directory containing the downloaded reference data must be specified with
`--jaffal_refBase`.
**JAFFAL annotation and genome files**
**Using alternative genome and annotation files**
The prepared JAFFAL reference files will look something like `hg38_chr20_genCode22.fa`. To enable JAFFAL to find these
files `--jaffal_genome` should be set to `hg38_chr20` and `--jaffal_annotation` to `genCode22`
These should be prepared as described
[here](https://github.com/Oshlack/JAFFA/wiki/FAQandTroubleshooting#how-can-i-generate-the-reference-files-for-a-non-supported-genome).
The resulting JAFFAL reference files will look something like `hg38_genCode22.fa`. The following options enable JAFFAL to find these
files:
`--jaffal_genome` optional (default: `hg38`)
`--jaffal_annotation` optional (default: `genCode22`)
__JAFFAL Notes__:
g++ must be installed. JAFFAL is not currently working on Mac M1 (osx-arm64 architecture). If there are no fusion transcripts
detected, the workflow will terminate with an error at the JAFFAL stage. If this happens,
skip the JAFFAL stage by omitting ` --jaffal_refBase`
__Note__: JAFFAL is not currently working on Mac M1 (osx-arm64 architecture).
### Differential Expression

5
download_jaffal_references.sh
View File

@ -1,5 +0,0 @@
#!/bin/sh
#Download the data. We should we move the data out of Figshare?
wget -O JAFFA_REFERENCE_FILES_HG38_GENCODE22.V2.tar.gz https://figshare.com/ndownloader/files/25410494
tar -zxvf JAFFA_REFERENCE_FILES_HG38_GENCODE22.V2.tar.gz