\nNumber of non-zero genes decrease with days, especially on day 7:
\nhttps://www.kaggle.com/code/alexandervc/mmscel-count-nonzero-genes-decrease-daily?scriptVersionId=106016829&cellId=12
\n
![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2262596%2F135f8e251a2f118a5fba038c2f2ea627%2FScreenshot%202022-09-19%20122528.png?generation=1663583141277576&alt=media\")
Consider features - counts of non-zero elements, sums of elemtns , for both RNA-seq and ATAC-seq data.
\nWe observe clearly dependence of these features on days.
Findings Briefly:
\n For RNA-seq part:
\n 1) it decreases with days
\n 2) it is correlated with proliferation activity (measured by standard genes signatures e.g. used in Scanpy/Seurat - Tirosh genes).
\n For ATAC-seq:
\n 1) counts of non-zeroes sharply grow at day 3, and little decrease at day 7
\n 2) sum decrease at day 3, and little increase at day 2
More findings can be found at notebooks:
\nhttps://www.kaggle.com/code/alexandervc/mmscel-count-nonzero-genes-decrease-daily
\n(for RNA-seq part - Multiome targets) and https://www.kaggle.com/code/alexandervc/mmscel-atac-seq-eda02-sums-counts-day-dependence (for ATAC-seq - Multiome features)
Figure 1:
\nNumber of non-zero genes decrease with days, especially on day 7:
\nhttps://www.kaggle.com/code/alexandervc/mmscel-count-nonzero-genes-decrease-daily?scriptVersionId=106016829&cellId=12
\n
Figure 2:
\nNumber of non-zero genes behave similar to proliferation (cell cycle activity) score:
\nhttps://www.kaggle.com/code/alexandervc/mmscel-count-nonzero-genes-decrease-daily?scriptVersionId=106016829&cellId=18
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2262596%2Fc8193394776c30aa1a60c81df7cb9d61%2FScreenshot%202022-09-19%20122611.png?generation=1663583190933281&alt=media\")
Figure3:
\nProliferation (cell cycle activity) score decreases with days:
\nhttps://www.kaggle.com/code/alexandervc/mmscel-count-nonzero-genes-decrease-daily?scriptVersionId=106016829&cellId=18
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2262596%2Fb07a131a3fb4279356b74167a0ae7912%2FScreenshot%202022-09-19%20122710.png?generation=1663583248479718&alt=media\")
Figure 4:
\nATAC seq counts of non-zeros
\nhttps://www.kaggle.com/code/alexandervc/mmscel-atac-seq-eda02-sums-counts-day-dependence?scriptVersionId=106056988&cellId=14
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2262596%2F1f831fe08ac607a7051926c02af8289d%2F__results___13_0.png?generation=1663620069848210&alt=media\")
Figure 5:
\nATAC seq sum of values
\nhttps://www.kaggle.com/code/alexandervc/mmscel-atac-seq-eda02-sums-counts-day-dependence?scriptVersionId=106056988&cellId=18
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2262596%2F739a4bd0d2d41ba1b7466ec89aa2816c%2F__results___17_0%20(3).png?generation=1663620162047539&alt=media\")
Consider features - counts of non-zero elements, sums of elemtns , for both RNA-seq and ATAC-seq data.
\nWe observe clearly dependence of these features on days.
Findings Briefly:
\n For RNA-seq part:
\n 1) it decreases with days
\n 2) it is correlated with proliferation activity (measured by standard genes signatures e.g. used in Scanpy/Seurat - Tirosh genes).
\n For ATAC-seq:
\n 1) counts of non-zeroes sharply grow at day 3, and little decrease at day 7
\n 2) sum decrease at day 3, and little increase at day 2
More findings can be found at notebooks:
\nhttps://www.kaggle.com/code/alexandervc/mmscel-count-nonzero-genes-decrease-daily
\n(for RNA-seq part - Multiome targets) and https://www.kaggle.com/code/alexandervc/mmscel-atac-seq-eda02-sums-counts-day-dependence (for ATAC-seq - Multiome features)
Figure 1:
\nNumber of non-zero genes decrease with days, especially on day 7:
\nhttps://www.kaggle.com/code/alexandervc/mmscel-count-nonzero-genes-decrease-daily?scriptVersionId=106016829&cellId=12
\n
Figure 2:
\nNumber of non-zero genes behave similar to proliferation (cell cycle activity) score:
\nhttps://www.kaggle.com/code/alexandervc/mmscel-count-nonzero-genes-decrease-daily?scriptVersionId=106016829&cellId=18
\n
Figure3:
\nProliferation (cell cycle activity) score decreases with days:
\nhttps://www.kaggle.com/code/alexandervc/mmscel-count-nonzero-genes-decrease-daily?scriptVersionId=106016829&cellId=18
\n
Figure 4:
\nATAC seq counts of non-zeros
\nhttps://www.kaggle.com/code/alexandervc/mmscel-atac-seq-eda02-sums-counts-day-dependence?scriptVersionId=106056988&cellId=14
\n
Figure 5:
\nATAC seq sum of values
\nhttps://www.kaggle.com/code/alexandervc/mmscel-atac-seq-eda02-sums-counts-day-dependence?scriptVersionId=106056988&cellId=18
\n
Even more simple feature - sum of all elements - shows similar behaviour.
\nThat a bit unexpected since sum of exponenets of these values are normalized to the same constant 1e6
Sum of genes expression for 13 genes placed in mitochondria chromosome - some days shows increase.
\nTypically increase of mitochondria genes is considered as indication towards \"badness\" of cells (near to dead).
![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2262596%2F2fa153f85dda33d790884dbb7f96ca53%2FScreenshot%202022-09-19%20135658.png?generation=1663588693844242&alt=media\")
Already mentioned - one donor was probably a woman, and women do not have \"Y\" chromosome.
\nAlso there is some decrease at day 7 of the expression - probably related to overall proliferation decrease.
![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2262596%2Fd5f8a9e62c8c0fba1a09ee239442e5bf%2FScreenshot%202022-09-19%20140204.png?generation=1663588947249778&alt=media\")
Updates including similar considerations for ATAC-seq are made.
", "rawMarkdown": "Updates including similar considerations for ATAC-seq are made.", "votes": null } ], "comments": [ { "id": 1945746, "author_name": "alexandervc", "author_url": "", "post_date": "09/19/2022 11:16:13", "content": "Even more simple feature - sum of all elements - shows similar behaviour.
\nThat a bit unexpected since sum of exponenets of these values are normalized to the same constant 1e6
Sum of genes expression for 13 genes placed in mitochondria chromosome - some days shows increase.
\nTypically increase of mitochondria genes is considered as indication towards \"badness\" of cells (near to dead).
Already mentioned - one donor was probably a woman, and women do not have \"Y\" chromosome.
\nAlso there is some decrease at day 7 of the expression - probably related to overall proliferation decrease.
Updates including similar considerations for ATAC-seq are made.
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1945695": "Consider features - counts of non-zero elements, sums of elemtns , for both RNA-seq and ATAC-seq data.\nWe observe clearly dependence of these features on days. \n\n**Findings Briefly:** \n For RNA-seq part:\n 1) it decreases with days \n 2) it is correlated with proliferation activity (measured by standard genes signatures e.g. used in Scanpy/Seurat - Tirosh genes).\n For ATAC-seq:\n 1) counts of non-zeroes sharply grow at day 3, and little decrease at day 7\n 2) sum decrease at day 3, and little increase at day 2\n\nMore findings can be found at notebooks: \nhttps://www.kaggle.com/code/alexandervc/mmscel-count-nonzero-genes-decrease-daily\n(for RNA-seq part - Multiome targets) and https://www.kaggle.com/code/alexandervc/mmscel-atac-seq-eda02-sums-counts-day-dependence (for ATAC-seq - Multiome features)\n\n-------------------------------\n\nFigure 1: \nNumber of non-zero genes decrease with days, especially on day 7:\nhttps://www.kaggle.com/code/alexandervc/mmscel-count-nonzero-genes-decrease-daily?scriptVersionId=106016829&cellId=12\n\n\nFigure 2: \nNumber of non-zero genes behave similar to proliferation (cell cycle activity) score:\nhttps://www.kaggle.com/code/alexandervc/mmscel-count-nonzero-genes-decrease-daily?scriptVersionId=106016829&cellId=18\n\n\nFigure3:\nProliferation (cell cycle activity) score decreases with days: \nhttps://www.kaggle.com/code/alexandervc/mmscel-count-nonzero-genes-decrease-daily?scriptVersionId=106016829&cellId=18\n\n\nFigure 4:\nATAC seq counts of non-zeros\nhttps://www.kaggle.com/code/alexandervc/mmscel-atac-seq-eda02-sums-counts-day-dependence?scriptVersionId=106056988&cellId=14\n\n\nFigure 5:\nATAC seq sum of values\nhttps://www.kaggle.com/code/alexandervc/mmscel-atac-seq-eda02-sums-counts-day-dependence?scriptVersionId=106056988&cellId=18\n.png?generation=1663620162047539&alt=media)", "1945746": "Even more simple feature - sum of all elements - shows similar behaviour.\nThat a bit unexpected since sum of exponenets of these values are normalized to the same constant 1e6\n\nhttps://www.kaggle.com/code/alexandervc/mmscel-count-nonzero-genes-decrease-daily?scriptVersionId=106019958&cellId=28\n", "1945790": "Sum of genes expression for 13 genes placed in mitochondria chromosome - some days shows increase.\nTypically increase of mitochondria genes is considered as indication towards \"badness\" of cells (near to dead). \n\n", "1945794": "Already mentioned - one donor was probably a woman, and women do not have \"Y\" chromosome.\nAlso there is some decrease at day 7 of the expression - probably related to overall proliferation decrease. \n\n", "1946492": "Updates including similar considerations for ATAC-seq are made." }, "source": "meta" }![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2262596%2F187ad055595458e9e6abde5c086028bd%2FScreenshot%202022-09-19%20131509.png?generation=1663586130564602&alt=media\")