README.Rmd

---
output: github_document
---

<!-- README.md is generated from README.Rmd. Please edit that file -->

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)
```

# UMIcountR

<!-- badges: start -->
<!-- badges: end -->

## Molecular Spikes

For information on obtaining molecular spikes, reference fasta and more, please visit the [molecular spikes GitHub repo](https://github.com/sandberg-lab/molecularSpikes).


## Installation


You can install UMIcountR from [GitHub](https://github.com/) with:

``` r
# install.packages("devtools")
devtools::install_github("cziegenhain/UMIcountR")
```
## Example

This is a basic example which shows you how to load and analyse some molecular spikes data:

```{r example}
library(UMIcountR)
## basic example code
#load reads from the provided example bam file (Smart-seq3 data)
bam_path <- system.file("extdata", "Smartseq3.TTACCTGCCAGATTCG.bam", package = "UMIcountR", mustWork = TRUE)

#in the case of the simple v1 molecular spike
spikedat <- extract_spike_dat(bam_path, match_seq_before_UMI = "GAGCCTGGGGGAACAGGTAGG", match_seq_after_UMI = "CTCGGAGGAGAAA")

#in the case of the complex molecular spikes set
data("molspike_barcodes_infos_fivep_final")
#spikedat <- extract_complex_spike_dat(bam_path, bc_df = spike_info, max_pattern_dist = 3)

```

After loading the data, we can see the data structure: 
```{r spikedat}
str(spikedat)
```


Next, we can run the filtering for overrespresented spUMIs:
```{r overrepresentation}
overrep <- get_overrepresented_spikes(spikedat, readcutoff = 75)
overrep$plots[[1]]
```


You could also apply directional-adjacency error correction to the Smart-seq3 UMI in the test data:
```{r ham}
spikedat[, UB_directional := return_corrected_umi(UX, editham = 1, collapse_mode = "adjacency_directional"), by = BC]
```

To downsample the copy number of the molecular spikes in your data to a relevant "expression level" of interest, you can use the following function:
```{r subs}
spikedat_mean100 <- subsample_recompute(spikedat, mu_nSpikeUMI = 100, threads = 4)
```


## Reference

Molecular spikes: a gold standard for single-cell RNA counting 
https://www.nature.com/articles/s41592-022-01446-x