---
title: "Quick Start Guide"
author: "Zaoqu Liu"
date: "`r Sys.Date()`"
output: 
  rmarkdown::html_vignette:
    toc: true
    toc_depth: 3
vignette: >
  %\VignetteIndexEntry{Quick Start Guide}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.width = 7,
  fig.height = 5,
  warning = FALSE,
  message = FALSE
)
```

## Introduction

**scTenifoldKnk** is an R package for performing *in-silico* knockout experiments using single-cell RNA sequencing data. It allows researchers to predict the effects of gene knockouts without performing actual wet-lab experiments.

### Key Features

- **Virtual Knockout**: Simulate gene knockouts computationally
- **Gene Regulatory Network**: Construct single-cell gene regulatory networks (scGRN)
- **Differential Regulation**: Identify genes affected by virtual knockout
- **High Performance**: C++ acceleration with Eigen library
- **Cross-Platform**: Works on macOS, Linux, and Windows

## Installation

```{r install, eval=FALSE}
# Install from GitHub
devtools::install_github("Zaoqu-Liu/scTenifoldKnk")
```

## Quick Example

### Load Package and Data

```{r load}
library(scTenifoldKnk)
library(Matrix)

# Load example data
data_path <- system.file("single-cell/example.csv", package = "scTenifoldKnk")
scRNAseq <- read.csv(data_path, row.names = 1)
scRNAseq <- as.matrix(scRNAseq)

# Check data dimensions
cat("Genes:", nrow(scRNAseq), "\n")
cat("Cells:", ncol(scRNAseq), "\n")
```

### Run Virtual Knockout Analysis

```{r run_analysis}
# Perform virtual knockout of gene G100
result <- scTenifoldKnk(
  countMatrix = scRNAseq,
  gKO = "G100",
  qc_minLSize = 0,      # Skip library size filter for demo data
  nc_nNet = 5,          # Number of networks
  nc_nCells = 100,      # Cells per network
  nc_nComp = 3,         # Principal components
  td_K = 3,             # Tensor rank
  verbose = TRUE
)
```

### View Results

```{r view_results}
# Top affected genes
head(result$diffRegulation, 10)
```

### Visualize Results

```{r visualize, fig.width=8, fig.height=6}
# Volcano-style plot
dr <- result$diffRegulation
dr$log10_padj <- -log10(dr$p.adj + 1e-300)
dr$significant <- dr$p.adj < 0.05

# Create plot
plot(dr$FC, dr$log10_padj,
     pch = 19,
     col = ifelse(dr$significant, "red", "gray60"),
     xlab = "Fold Change (FC)",
     ylab = "-log10(adjusted p-value)",
     main = "Virtual Knockout Effect: G100",
     cex = 0.8)

# Highlight knockout gene
ko_idx <- which(dr$gene == "G100")
points(dr$FC[ko_idx], dr$log10_padj[ko_idx], 
       pch = 17, col = "blue", cex = 2)

# Add legend
legend("topright", 
       legend = c("Significant (FDR<0.05)", "Not significant", "Knockout gene"),
       col = c("red", "gray60", "blue"),
       pch = c(19, 19, 17),
       bty = "n")

# Add threshold line
abline(h = -log10(0.05), lty = 2, col = "gray40")
```

## Output Structure

The `scTenifoldKnk` function returns a list with the following components:

| Component | Description |
|-----------|-------------|
| `tensorNetworks$WT` | Wild-type tensor network |
| `tensorNetworks$KO` | Knockout tensor network |
| `manifoldAlignment` | Manifold alignment matrix |
| `diffRegulation` | Differential regulation results |

### Differential Regulation Table

```{r dr_table}
knitr::kable(head(result$diffRegulation), 
             caption = "Top Differentially Regulated Genes",
             digits = 4)
```

## Next Steps

- See the [Algorithm Theory](algorithm-theory.html) vignette for detailed methodology
- See the [Visualization Guide](visualization.html) for advanced plotting
- See the [Advanced Usage](advanced-usage.html) for parameter tuning

## Session Info

```{r session}
sessionInfo()
```