Last updated: 2018-09-14

workflowr checks: (Click a bullet for more information)

• ✔ R Markdown file: up-to-date

  Great! Since the R Markdown file has been committed to the Git repository, you know the exact version of the code that produced these results.

• ✔ Repository version: 86bd1c0

  Great! You are using Git for version control. Tracking code development and connecting the code version to the results is critical for reproducibility. The version displayed above was the version of the Git repository at the time these results were generated.
  
  Note that you need to be careful to ensure that all relevant files for the analysis have been committed to Git prior to generating the results (you can use wflow_publish or wflow_git_commit). workflowr only checks the R Markdown file, but you know if there are other scripts or data files that it depends on. Below is the status of the Git repository when the results were generated:

  
  Ignored files:
      Ignored:    .Rhistory
      Ignored:    .Rproj.user/
      Ignored:    R/.Rhistory
      Ignored:    analysis/.Rhistory
      Ignored:    analysis/pipeline/.Rhistory
  
  Untracked files:
      Untracked:  ..gif
      Untracked:  .DS_Store
      Untracked:  R/.DS_Store
      Untracked:  analysis/.DS_Store
      Untracked:  analysis/normalization_test.R
      Untracked:  analysis/pipeline/0_dropseq/
      Untracked:  analysis/pipeline/1_10X/
      Untracked:  analysis/pipeline/2_zeisel/
      Untracked:  analysis/pipeline/3_smallsets/
      Untracked:  analysis/slsl_10x.Rdata
      Untracked:  analysis/slsl_dropseq.Rdata
      Untracked:  analysis/writeup/bibliography.bib
      Untracked:  analysis/writeup/draft1.aux
      Untracked:  analysis/writeup/draft1.bbl
      Untracked:  analysis/writeup/draft1.blg
      Untracked:  analysis/writeup/draft1.log
      Untracked:  analysis/writeup/draft1.out
      Untracked:  analysis/writeup/draft1.pdf
      Untracked:  analysis/writeup/draft1.synctex.gz
      Untracked:  analysis/writeup/draft1.tex
      Untracked:  analysis/writeup/jabbrv-ltwa-all.ldf
      Untracked:  analysis/writeup/jabbrv-ltwa-en.ldf
      Untracked:  analysis/writeup/jabbrv.sty
      Untracked:  analysis/writeup/naturemag-doi.bst
      Untracked:  analysis/writeup/wlscirep.cls
      Untracked:  data/unnecessary_in_building/
      Untracked:  docs/figure/10x_labeled.Rmd/
      Untracked:  docs/figure/example_10x.Rmd/.DS_Store
      Untracked:  docs/figure/large.Rmd/
      Untracked:  docs/figure/small.Rmd/
      Untracked:  dropseq_heatmap.pdf
      Untracked:  man/correct_detection_rate.Rd
  
  Unstaged changes:
      Modified:   NAMESPACE
      Modified:   R/SLSL.R
      Modified:   R/correct_detection_rate.R
      Modified:   R/find_markers.R
      Modified:   analysis/pipeline/.DS_Store
      Modified:   analysis/writeup/.DS_Store
      Modified:   data/.DS_Store
      Modified:   docs/figure/.DS_Store
      Deleted:    man/correct.detection.rate.Rd
  
  

  Note that any generated files, e.g. HTML, png, CSS, etc., are not included in this status report because it is ok for generated content to have uncommitted changes.

Expand here to see past versions:

File	Version	Author	Date	Message
html	1333c7f	tk382	2018-08-30	Build site.
Rmd	64d04da	tk382	2018-08-29	add toc
html	2a3ec97	tk382	2018-08-26	Build site.
html	efd32b3	tk382	2018-07-23	Build site.
html	a20c598	tk382	2018-07-21	Build site.
html	5787a2e	tk382	2018-07-17	Build site.
html	410c051	tk382	2018-07-17	Build site.
Rmd	f6c5b07	tk382	2018-07-11	add website
html	f6c5b07	tk382	2018-07-11	add website

---

SLSL (Scaled Lasso Similarity Learning) is an unsupervised learning tool for clustering single cells using their expression level data. SLSL constructs many similarity matrices using different distance measures (Euclidean, Pearson, and Spearman), and different kernel parameters to account for the possible nonlinear structure. Then, SLSL employs scaled lasso to find optimal weights on each of those similarity matrices and infer one final sparse similarity matrix. This procedure does not have rank constraint unlike many popular unsupervised learning tools, because imposing such inflexible structure does not guarantee higher accuracy. Instead, we employ network diffusion to make the final similarity matrix closer to the block diagonal matrix. This is in effect equivalent to shrinking the less important eigenvalues. Lastly, SLSL sequentially performs dimension reduction using tSNE and kmeans for the final clustering result.

---

This reproducible R Markdown analysis was created with workflowr 1.1.1

---