https://github.com/andrewdavidsmith/transferase
The transferase system for retrieving methylome data from methbase
https://github.com/andrewdavidsmith/transferase
bisulfite bisulfite-sequencing dna-methylation epigenetics epigenomics methyl-seq methylation methylation-analysis methylome wgbs
Last synced: about 1 month ago
JSON representation
The transferase system for retrieving methylome data from methbase
- Host: GitHub
- URL: https://github.com/andrewdavidsmith/transferase
- Owner: andrewdavidsmith
- License: mit
- Created: 2024-10-06T19:01:30.000Z (8 months ago)
- Default Branch: main
- Last Pushed: 2025-04-03T19:04:20.000Z (about 2 months ago)
- Last Synced: 2025-04-03T20:22:18.791Z (about 2 months ago)
- Topics: bisulfite, bisulfite-sequencing, dna-methylation, epigenetics, epigenomics, methyl-seq, methylation, methylation-analysis, methylome, wgbs
- Language: C++
- Homepage:
- Size: 3.38 MB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- License: LICENSE
Awesome Lists containing this project
README
# Transferase
[What's New](#whats-new)
The MethBase2 database currently has roughly 13,500 *high-quality* methylomes
for vertebrate species, mostly split equally between human and mouse. For most
epigenomics scientists, answering important questions requires identifying
relevant data sets and obtaining methylation levels summarized at relavant
parts of the genome. Transferase provides fast access to MethBase2 via
queries made up of two parts: a set of genomic intervals and one or more
methylomes of interest. Queries return methylation levels in each interval for
each specified methylome. There is no need to download entire methylomes
because the server can quickly respond with the methylation levels in query
regions -- tens of thousands of intervals in a single query, with responses as
fast as a few seconds over regular home internet. In some situations the
transferse server can respond faster than you could load the data files if had
them locally on your laptop or workstation.
- As of v0.5.0, the server open to the public. I don't yet know how well the
server can stand up to heavy traffic -- even from a small number of users
with big queries. So I might need to put restrictions on the sizes of
queries. As of now, it's quite open.
- The current version (v0.6.0) is still in an early development stage,
although many improvements have been made. The command line tools are
maturing, but hasn't seen enough use to know if there are any serious bugs
lurking. This is especially true for the Python bindings, and even more for
the R package (Rxfr).
- I appreciate feedback. You can open an issue on github or contact me
(Andrew) directly.
- If you have a lot of methylome data, whether it's from whole-genome
bisulfite sequencing, or from Nanopore, you can use transferase locally to
do very fast data analysis.
- Target systems are Mac and Linux. As of v0.6.0, there are binary releases
that should work on any Linux and any Mac machine. Only two users have
reported that the binaries don't work on their systems.
- Target platforms are command line, Python and R. The R API is available as
of v0.6.0.
- The Python package needs Python 3.13, but otherwise will run on almost any
Linux machine. The R package needs at least R version 4.4.
## What's new
* The transferase Python package, pyxfr, can now be installed using
[pip](https://pypi.org/project/pyxfr/0.6.0) with Python >= 3.12 on Linux
and macOS (at least Ventura):
```console
pip install pyxfr
```
## Documentation
The instructions for using transferase on the command line are in the
`docs/command_line.md` file and very basic docs for the Python API are in
`docs/python.md`. There is now built-in Python documentation (e.g., using
`help(transferase)` within Python), and similarly for R.
## Installing transferase
### Building from source
Detailed (maybe too detailed) instructions for building transferase from
source can be found in the `docs/building.md` file in this repo.
### Linux
You can find all the available installers from "releases" page (assuming you
are reading this on github.com). You can download a shell installer
[here](https://github.com/andrewdavidsmith/transferase/releases/download/v0.6.0/transferase-0.6.0-Linux.sh).
Then run the downloaded installer (likely you want to first install it beneath
your home dir):
```console
sh transferase-0.6.0-Linux.sh --prefix=${PREFIX}
```
This will prompt you to accept the license, and then it will install the
transferase binaries in `${PREFIX}/bin`, along with some config files in
`${PREFIX}/share`. If you want to install it system-wide, and have admin
privileges, you can do:
```console
sh transferase-0.6.0-Linux.sh --prefix=/usr/local
```
If you are on Debian or Ubuntu, and have admin privileges, you can use the
Debian package and then transferase will be tracked in the package management
system. Similarly, there is an RPM for Linux distributions that use those. You
can get these from the "releases" page. For Ubuntu and Debian, install like
this:
```console
sudo dpkg -i ./transferase-0.6.0-Linux.deb
```
For Red Hat, Fedora or SUSE, install like this:
```console
sudo rpm -i ./transferase-0.6.0-Linux.rpm
```
### Mac
For Mac two architectures are supported: the older "x86_64" and the the arm64
("Apple Silicon"). If you are using an older Mac, then the binaries built on
macOS 13 should work for you. If you are using a newer Mac, then the binaries
built on macOS 14 or 15 should work for you. I think this should be true
regardless of which version of macOS you are actually using -- as long as you
have the right hardware architecture. However, I can't test all systems in all
combinations, so feedback is appreciated.
Curently only a shell installer is available. You can download the installer
matching your version at the releases page on github. Assuming it's macos-15
and ARM64, then do this:
```console
sh transferase-0.6.0-Linux.sh --prefix=${PREFIX}
```
This will prompt you to accept the license, and then it will install the
transferase binaries in `${PREFIX}/bin`, along with some config files in
`${PREFIX}/share`. For v0.6.0, be careful with the bash completions file on
macos -- it's still a bit unstable.
## Installing the Python package
The transferase Python package, pyxfr, can now be installed using
[pip](https://pypi.org/project/pyxfr/0.6.0) with Python >= 3.12 on Linux
and macOS (at least Ventura):
```console
python3 -m venv .venv
. .venv/bin/activate
pip install pyxfr
# To test -- if you see output, it worked!
python3 -c "from pyxfr import *; help(pyxfr)"
```
## Installing the R API
Please see the instructions in
[`docs/installing_Rxfr.md`](docs/installing_Rxfr.md).