cytodata / awesome-cytodata Goto Github PK

… or would that be considered "opinion presented as fact"?

e.g. I was maintaining this list a few years ago
https://paperpile.com/app/shared/i40RyD which I considered a getting started reading list.

At present, we have only cytominer.

Here are others I could think of

• scanpy
• PurifyR

We now also have

• pycytominer

Hi! While updating the repository, I came across the following references, for which I was unsure if they are in the scope, if they are too redundant with the content already included or I was simply unable to judge if they should be included or not. I believe most could be useful for the community but are not necessary involving image-based profiling tasks directly.

Functionally-Relevant Morphological Profiling: A Tool to Assess Cellular Heterogeneity
Morphological profiling of small molecules
Label-free cell cycle analysis for high-throughput imaging flow cytometry
Automated analysis of high-content microscopy data with deep learning
Time series modeling of live-cell shape dynamics for image-based phenotypic profiling
Domain-invariant features for mechanism of action prediction in a multi-cell-line drug screen
Classifying and segmenting microscopy images with deep multiple instance learning
KCML: a machine‐learning framework for inference of multi‐scale gene functions from genetic perturbation screens
Visualizing quantitative microscopy data: History and challenges
Transfer Learning with Deep Convolutional Neural Networks for Classifying Cellular Morphological Changes
High-content phenotypic and pathway profiling to advance drug discovery in diseases of unmet need
RefCell: multi-dimensional analysis of image-based high-throughput screens based on ‘typical cells’
Pooled CRISPR screens with imaging on microraft arrays reveals stress granule-regulatory factors
Label-free prediction of three-dimensional fluorescence images from transmitted-light microscopy
High-content analysis screening for cell cycle regulators using arrayed synthetic crRNA libraries
Deep learning is combined with massive-scale citizen science to improve large-scale image classification
Content-aware image restoration: pushing the limits of fluorescence microscopy
Optical Pooled Screens in Human Cells

If you think some of those articles are awesome for image-based profiling of biological phenotypes, feel free to submit a PR!

Let's use this issue to collect a list of batch correction methods from both, image-based profiling, as well as related profiling modalities.

From what I can tell, the three entries (BBBC, IDR, RxRx1) in the datasets resource section include only raw images.

Are there any known examples of resources that provide actual profiles (in addition to raw images)?

One of the major reasons for success in RNA space is that processed matrices are made available alongside raw data. This enables really quick secondary analyses. I haven't really seen much of this for profiling data.

We have an option of adding our awesome-cytodata collection to the awesome stuff curation page (https://github.com/sindresorhus/awesome).

They have quite a complicated contribution guideline! (with over 100K stars, this is not that surprising!)

We should consider doing this. It will

1. increase visibility
2. solidify what does and what does not belong in this repo

If we decide to add to the official awesome list, I think we should restructure the repo a bit.

• Do not add publications as issues (this is a good idea that we should do anyway in a separate repo)
• Only maintain "awesome" publication resources.
  • This probably only includes comprehensive review articles.
  • And maybe some landmark papers?
• All else stays the same (and should be added upon). i.e. data sets, software, etc.