> For the complete documentation index, see [llms.txt](https://sxu99.gitbook.io/ssblazer/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://sxu99.gitbook.io/ssblazer/getting-started.md).

# Getting Started

## Overview

SSBlazer represents a pioneering effort to harness deep learning techniques for the prediction of Single Strand Break (SSB) sites. It utilizes genome-wide single-strand break sequencing data with single-nucleotide resolution to make these predictions. For more information, to download the tool, or to contribute, visit the [SSBlazer GitHub repository](https://github.com/sxu99/ssblazer). For a web-based interface that enables easy use of this tool, check out the [SSBlazer Web Server](https://proj.cse.cuhk.edu.hk/aihlab/ssblazer/). 

<figure><img src="/files/4gBvpFkEv34xOxAXG7g6" alt=""><figcaption><p>Pipeline of SSBlazer</p></figcaption></figure>

## Operating Instructions

SSBlazer is designed to perform three primary functions:

1. **Prediction of DNA Single Strand Break Sites**\
   SSBlazer can accurately predict the probability of single strand breaks at each nucleotide position within a DNA sequence provided in FASTA format. ([Manual](/ssblazer/use-cases/break-sites-prediction.md), [Web Server](https://proj.cse.cuhk.edu.hk/aihlab/ssblazer/#/home))
2. **Point Mutation Analysis:**\
   The software can also predict the impact of point mutations on nearby DNA break probability. ([Manual](/ssblazer/use-cases/mutation-analysis.md), [Web Server](https://proj.cse.cuhk.edu.hk/aihlab/ssblazer/#/pred_snp/))
3. **Training a New Model:**\
   Beyond SSB predictions, SSBlazer's framework allows for training on different lesion types, such as double-strand breaks, by supplying a dataset specific to the desired lesion type. ([Manual](/ssblazer/use-cases/train-a-new-model.md))

{% hint style="info" %}
For optimal performance, input lengths are limited in the web server. Please use our local software version for larger tasks.
{% endhint %}

## Environment Setup

To get started with SSBlazer, we need to set up the environment correctly.

#### Step 1: Clone the Repository

First, download the repository by cloning it from the GitHub repository.

```sh
git clone https://github.com/sxu99/ssblazer.git
```

Once cloned, you need to navigate into the `ssblazer` directory.

```sh
cd ./ssblazer
```

#### Step 2: Create the Environment

Next, you will need to create a new environment. We use the `conda` package manager for this. If you don't have `conda` installed, you can download it from the [official site](https://docs.conda.io/en/latest/miniconda.html).

With `conda` installed, create the environment using the following command:

```sh
conda env create -f environment.yml
```

#### Step 3: Activate the Environment

Once the environment is created, you can activate it using the following command:

```sh
conda activate ssblazer
```

After these steps, your environment is set up and ready to go!

## Predict SSB Sites from Sequence

With SSBlazer, you can make predictions either from a local fasta file or using our online tool. You can use a local fasta file to make predictions. The following command uses a test fasta file and a batch size of 128:

```sh
python prediction.py --file ./test.fa --batchsize 128
```

After running the prediction script, the results will be saved in a `.bed` file format. This file, `result.bed`, contains the predicted SSB sites.

## Contact Us

This software is developed by [Sheng Xu](https://sxu99.github.io/), [Junkang Wei](https://tikeng.github.io/) and [Yu Li](https://liyu95.com/). If you have any questions or need assistance with SSBlazer, you can contact Yu Li directly at `liyu[AT]cse.cuhk.edu.hk`. For bug reporting, feature requests, or general questions about the tool, you may also raise an issue on the [SSBlazer GitHub repository](https://github.com/sxu99/ssblazer).&#x20;

## Citation

If you use SSBlazer in your work, please cite the publication:

* Xu S, Wei J, Li Y. Genome-wide nucleotide-resolution model of single-strand break site reveals species evolutionary hierarchy\[J]. arXiv preprint arXiv:2208.09813, 2022.