input file for score

[BenxiaHu]

Hello, what is the input file (--coord TEXT Path to stripe coordinate table) for stripenn score? is it the result_filtered.tsv? Best,

[ysora]

Yes, that's right!

Get Outlook for Androidhttps://aka.ms/AAb9ysg

---

From: bxhu @.> Sent: Monday, September 26, 2022 3:57:43 PM To: ysora/stripenn @.> Cc: Subscribed @.***> Subject: [External] [ysora/stripenn] input file for score (Issue #6)

Hello, what is the input file (--coord TEXT Path to stripe coordinate table) for stripenn score? is it the result_filtered.tsv? Best,

— Reply to this email directly, view it on GitHubhttps://github.com/ysora/stripenn/issues/6, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ABZHMJVMJVPWDIICWUUHU63WAH53PANCNFSM6AAAAAAQWD4WTI. You are receiving this because you are subscribed to this thread.Message ID: @.***>

[BenxiaHu]

I am still confused of how to run stripenn score. For example, I have case and control samples, and called stripes from both smaples. How to call differential stripes betweeen case and control samples? Another feature required: would you like to tell me how to annotate stripes?

[ysora]

1. Concatenate control and test data (row bind). Let say this combined data as A.
2. Run stripenn score of A for control cool file.
3. Run stripenn score of A for test cool file. Then you will get stripiness of all stripes for both control and test conditions.

Next you can compare stripiness of a certain stripe for control and test conditions.

In the stripenn paper, I counted a stripe as test specific if its stripiness is larger than 2 in test and less than 0 in control.

This is just an example and you can use any approach to find differential stripes.

Thanks, Sora

Get Outlook for Androidhttps://aka.ms/AAb9ysg

---

From: bxhu @.> Sent: Tuesday, September 27, 2022 11:53:14 AM To: ysora/stripenn @.> Cc: Yoon, Sora @.>; Comment @.> Subject: [External] Re: [ysora/stripenn] input file for score (Issue #6)

I am still confused of how to run stripenn score. For example, I have case and control samples, and called stripes from both smaples. How to call differential stripes betweeen case and control samples? Another feature required: would you like to tell me how to annotate stripes?

— Reply to this email directly, view it on GitHubhttps://github.com/ysora/stripenn/issues/6#issuecomment-1259705957, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ABZHMJQJSV3DHVULQRRKGBDWAMJ6VANCNFSM6AAAAAAQWD4WTI. You are receiving this because you commented.Message ID: @.***>

[ysora]

Could you be more specific in stripe annotation?

Get Outlook for Androidhttps://aka.ms/AAb9ysg

---

From: bxhu @.> Sent: Tuesday, September 27, 2022 11:53:14 AM To: ysora/stripenn @.> Cc: Yoon, Sora @.>; Comment @.> Subject: [External] Re: [ysora/stripenn] input file for score (Issue #6)

I am still confused of how to run stripenn score. For example, I have case and control samples, and called stripes from both smaples. How to call differential stripes betweeen case and control samples? Another feature required: would you like to tell me how to annotate stripes?

— Reply to this email directly, view it on GitHubhttps://github.com/ysora/stripenn/issues/6#issuecomment-1259705957, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ABZHMJQJSV3DHVULQRRKGBDWAMJ6VANCNFSM6AAAAAAQWD4WTI. You are receiving this because you commented.Message ID: @.***>

[BenxiaHu]

sorry for the late reply.

for example, chr1:1000-20000 is a stripe, how to predict it functions or tragte genes?

[ysora]

I am not sure if I understood your question correctly. Stripe is an architectural structure that one anchor interacts with contiguous loci. Among the interacting genomic loci, some regions are enriched with enhancer marks (H3K27ac). And I found that those regions are usually highly expressed... so I'd like to say the stripe can be annotated by that region (or it can be simple gene). Does that make sense? Thanks,

Sora

[Ariunaa5000]

1. Concatenate control and test data (row bind). Let say this combined data as A. 2. Run stripenn score of A for control cool file. 3. Run stripenn score of A for test cool file. Then you will get stripiness of all stripes for both control and test conditions. Next you can compare stripiness of a certain stripe for control and test conditions. In the stripenn paper, I counted a stripe as test specific if its stripiness is larger than 2 in test and less than 0 in control. This is just an example and you can use any approach to find differential stripes. Thanks, Sora Get Outlook for Androidhttps://aka.ms/AAb9ysg … ____ From: bxhu @.> Sent: Tuesday, September 27, 2022 11:53:14 AM To: ysora/stripenn @.> Cc: Yoon, Sora @.>; Comment @.> Subject: [External] Re: [ysora/stripenn] input file for score (Issue #6) I am still confused of how to run stripenn score. For example, I have case and control samples, and called stripes from both smaples. How to call differential stripes betweeen case and control samples? Another feature required: would you like to tell me how to annotate stripes? — Reply to this email directly, view it on GitHub<#6 (comment)>, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ABZHMJQJSV3DHVULQRRKGBDWAMJ6VANCNFSM6AAAAAAQWD4WTI. You are receiving this because you commented.Message ID: @.***>

Hi again Sora,

I have follow-up questions on interpreting the score function output.

• For a stripe to be considered specific to a test condition, does it have to come from the initial stripes called in test (as in if we say first half rows of A came from control and second half rows came from test). Can you explain the rationale of combining the control and test data before calling score function? That might help me understand it better.
• When you are using Stripiness > 2 cut off, are you using Stripiness or Stripiness_added column? What does "_added" columns refer to? Are they added in the sense that it was added after running score or are you adding different stripiness values to get that one?

Many thanks for your time.

[ysora]

Hi Ariunaa,

Let say we are doing differential stripe test for two conditions. It can be done in multiple ways. First, we can merge stripes from two conditions (here, we can remove the exact same stripes by using 'uniq' command), and then see their chromatin interaction differences in control and test data by running score function twice (one for control, and the other for the test). Second, we can extract stripes from only test data, and then get its scores for control Hi-C data. You can choose any method that is reasonable!

For your second question, '_added' column is added after you run the score function based on your new input .(m)cool file. i.e., the _added columns are whatever calculated based on the new input data. So, when I used Stripiness > 2 cutoff, I think I used the second 'Stripiness_added' column (since there will be two Stripiness_added columns, one for control and the other for the test).

Let me know if it is unclear.

Thanks, Sora

---

From: Ariunaa Bayanjargal @.> Sent: Thursday, August 24, 2023 2:38 PM To: ysora/stripenn @.> Cc: Yoon, Sora @.>; State change @.> Subject: [External] Re: [ysora/stripenn] input file for score (Issue #6)

1. Concatenate control and test data (row bind). Let say this combined data as A. 2. Run stripenn score of A for control cool file. 3. Run stripenn score of A for test cool file. Then you will get stripiness of all stripes for both control and test conditions. Next you can compare stripiness of a certain stripe for control and test conditions. In the stripenn paper, I counted a stripe as test specific if its stripiness is larger than 2 in test and less than 0 in control. This is just an example and you can use any approach to find differential stripes. Thanks, Sora Get Outlook for Androidhttps://aka.ms/AAb9ysg … ____ From: bxhu @.> Sent: Tuesday, September 27, 2022 11:53:14 AM To: ysora/stripenn @.> Cc: Yoon, Sora @.>; Comment @.> Subject: [External] Re: [ysora/stripenn] input file for score (Issue #6https://github.com/ysora/stripenn/issues/6) I am still confused of how to run stripenn score. For example, I have case and control samples, and called stripes from both smaples. How to call differential stripes betweeen case and control samples? Another feature required: would you like to tell me how to annotate stripes? — Reply to this email directly, view it on GitHub<#6 (comment)https://github.com/ysora/stripenn/issues/6#issuecomment-1259705957>, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ABZHMJQJSV3DHVULQRRKGBDWAMJ6VANCNFSM6AAAAAAQWD4WTI. You are receiving this because you commented.Message ID: @.***>

Hi again Sora,

I have follow-up questions on interpreting the score function output.

• For a stripe to be considered specific to a test condition, does it have to come from the initial stripes called in test (as in if we say first half rows of A came from control and second half rows came from test). Can you explain the rationale of combining the control and test data before calling score function? That might help me understand it better.
• When you are using Stripiness > 2 cut off, are you using Stripiness or Stripiness_added column? What does "_added" columns refer to? Are they added in the sense that it was added after running score or are you adding different stripiness values to get that one?

Many thanks for your time.

— Reply to this email directly, view it on GitHubhttps://github.com/ysora/stripenn/issues/6#issuecomment-1692223241, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ABZHMJXYWF2PIICAXDB3XFLXW6NTFANCNFSM6AAAAAAQWD4WTI. You are receiving this because you modified the open/close state.Message ID: @.***>