search

lewandowskilab / PVF_Manuscript

Supplementary data for the "Perivascular fibroblast manuscript"
2 stars 0 forks source link

different directionality in Figure 3? #1

Open nmra-ajaffe opened 3 years ago

nmra-ajaffe commented 3 years ago

Super impressive paper and efforts, and I applaud your group for making the data underlying your clinical associations available. However, it seemed like the directionality of SPP1 was the opposite in the Rdata object compared to Figure 3a - is there some other form of normalization necessary? 

load("2020_02_Combined_Cohort_Dataframes copy.Rdata")

## recode
cohorts$dx = cohorts$class
cohorts$dx[cohorts$dx == "Neuro control"] = "Control"
cohorts$dx[cohorts$dx == "ALS"] = "Case"
cohorts$dx = droplevels(cohorts$dx)
table(cohorts$dx)
# Case Control 
# 479     422 

t.test(RawB.53 ~ dx, data= cohorts)
# data:  RawB.53 by dx
# t = -4.6369, df = 673.65, p-value = 4.249e-06
# alternative hypothesis: true difference in means is not equal to 0
# 95 percent confidence interval:
#   -0.1500042 -0.0607571
# sample estimates:
#   mean in group Case mean in group Control 
# 7.358855              7.464236
lewandowskilab commented 3 years ago

Hello Andrew, Thanks for your insight. I will look into the matter and get back to you. Best wishes

Sebastian

Sebastian Lewandowski PhD

Karolinska Institutet, CMM, L8:04 Karolinska vägen 8A SE-171 64 Solna Twitter @LewandowskiSA

From: Andrew Jaffe @.***> Sent: Friday, April 16, 2021 5:47 AM To: lewandowskilab/PVF_Manuscript Cc: Subscribed Subject: [lewandowskilab/PVF_Manuscript] different directionality in Figure 3? (#1)

Super impressive paper and efforts, and I applaud your group for making the data underlying your clinical associations available. However, it seemed like the directionality of SPP1 was the opposite in the Rdata object compared to Figure 3a - is there some other form of normalization necessary?

load("2020_02_Combined_Cohort_Dataframes copy.Rdata")

recode

cohorts$dx = cohorts$class cohorts$dx[cohorts$dx == "Neuro control"] = "Control" cohorts$dx[cohorts$dx == "ALS"] = "Case" cohorts$dx = droplevels(cohorts$dx) table(cohorts$dx)

Case Control

479 422

t.test(RawB.53 ~ dx, data= cohorts)

data: RawB.53 by dx

t = -4.6369, df = 673.65, p-value = 4.249e-06

alternative hypothesis: true difference in means is not equal to 0

95 percent confidence interval:

-0.1500042 -0.0607571

sample estimates:

mean in group Case mean in group Control

7.358855 7.464236

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Flewandowskilab%2FPVF_Manuscript%2Fissues%2F1&data=04%7C01%7Csebastian.lewandowski%40ki.se%7Cfd37dc44df0c44f31b6308d9008a6b8d%7Cbff7eef1cf4b4f32be3da1dda043c05d%7C0%7C0%7C637541416780869394%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=AAuph0C0ewWC9Ibs5Q7H%2FpiJCfXuy2qX9EGNgWkyNU0%3D&reserved=0, or unsubscribehttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAMHSJKB6WX7NRBA5677NVXTTI6XOVANCNFSM43AWVHMQ&data=04%7C01%7Csebastian.lewandowski%40ki.se%7Cfd37dc44df0c44f31b6308d9008a6b8d%7Cbff7eef1cf4b4f32be3da1dda043c05d%7C0%7C0%7C637541416780869394%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=NNq%2F%2B1Is9vPfGdiIa9yPtVtVTew%2BuFF4qkDqbgfC21U%3D&reserved=0.

När du skickar e-post till Karolinska Institutet (KI) innebär detta att KI kommer att behandla dina personuppgifter. Här finns information om hur KI behandlar personuppgifterhttps://ki.se/medarbetare/integritetsskyddspolicy.

Sending email to Karolinska Institutet (KI) will result in KI processing your personal data. You can read more about KI’s processing of personal data herehttps://ki.se/en/staff/data-protection-policy.

nmra-ajaffe commented 3 years ago

Looking at this a bit more, I wonder if the README should state that RawB.94 is actually SPP1?

From README:

RawB.53, RawB.123 and RawB.94
These are the log normalized MFI values where;
  *   53 is SPP1
  *   123 is COL6A1
  *   94 is NEFL

Distributions of each variable by Dx image

Corresponding Figure 3a: image

Thanks!

From: lewandowskilab @.> Sent: Friday, April 16, 2021 5:04 AM To: lewandowskilab/PVF_Manuscript @.> Cc: Andrew Jaffe @.>; Author @.> Subject: Re: [lewandowskilab/PVF_Manuscript] different directionality in Figure 3? (#1)

Hello Andrew, Thanks for your insight. I will look into the matter and get back to you. Best wishes

Sebastian

Sebastian Lewandowski PhD

Karolinska Institutet, CMM, L8:04 Karolinska vägen 8A SE-171 64 Solna Twitter @LewandowskiSA

From: Andrew Jaffe @.<mailto:@.>> Sent: Friday, April 16, 2021 5:47 AM To: lewandowskilab/PVF_Manuscript Cc: Subscribed Subject: [lewandowskilab/PVF_Manuscript] different directionality in Figure 3? (#1)

Super impressive paper and efforts, and I applaud your group for making the data underlying your clinical associations available. However, it seemed like the directionality of SPP1 was the opposite in the Rdata object compared to Figure 3a - is there some other form of normalization necessary?

load("2020_02_Combined_Cohort_Dataframes copy.Rdata")

recode

cohorts$dx = cohorts$class cohorts$dx[cohorts$dx == "Neuro control"] = "Control" cohorts$dx[cohorts$dx == "ALS"] = "Case" cohorts$dx = droplevels(cohorts$dx) table(cohorts$dx)

Case Control

479 422

t.test(RawB.53 ~ dx, data= cohorts)

data: RawB.53 by dx

t = -4.6369, df = 673.65, p-value = 4.249e-06

alternative hypothesis: true difference in means is not equal to 0

95 percent confidence interval:

-0.1500042 -0.0607571

sample estimates:

mean in group Case mean in group Control

7.358855 7.464236

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Flewandowskilab%2FPVF_Manuscript%2Fissues%2F1&data=04%7C01%7Csebastian.lewandowski%40ki.se%7Cfd37dc44df0c44f31b6308d9008a6b8d%7Cbff7eef1cf4b4f32be3da1dda043c05d%7C0%7C0%7C637541416780869394%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=AAuph0C0ewWC9Ibs5Q7H%2FpiJCfXuy2qX9EGNgWkyNU0%3D&reserved=0, or unsubscribehttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAMHSJKB6WX7NRBA5677NVXTTI6XOVANCNFSM43AWVHMQ&data=04%7C01%7Csebastian.lewandowski%40ki.se%7Cfd37dc44df0c44f31b6308d9008a6b8d%7Cbff7eef1cf4b4f32be3da1dda043c05d%7C0%7C0%7C637541416780869394%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=NNq%2F%2B1Is9vPfGdiIa9yPtVtVTew%2BuFF4qkDqbgfC21U%3D&reserved=0.

När du skickar e-post till Karolinska Institutet (KI) innebär detta att KI kommer att behandla dina personuppgifter. Här finns information om hur KI behandlar personuppgifterhttps://ki.se/medarbetare/integritetsskyddspolicy.

Sending email to Karolinska Institutet (KI) will result in KI processing your personal data. You can read more about KI’s processing of personal data herehttps://ki.se/en/staff/data-protection-policy.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/lewandowskilab/PVF_Manuscript/issues/1#issuecomment-821033571, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ARC3BQRJERLCRPGM6JUHAVTTI74PZANCNFSM43AWVHMQ.

lewandowskilab commented 3 years ago

Hello Andrew Thanks for pointing out the difference in SPP1 data on our GitHub.

I guess that by "directionality" you mean the relationship of datapoint distributions and their medians between ALS and Controls. The apparent difference between Figure 3a and the plot from RawB.53 is there for the following reason:

RawB.53 represents the "Validation antibody" for SPP1 (HPA005562). (you can see a graph representing this antibody in ALS vs. controls in Extended data figure 9A and it is very similar to yours)

Figure 3A represents the "primary" SPP1 antibody (HPA027541) (RawB.128)

We did not aim to claim that SPP1 increase will distinguish ALS cases from controls and therefore did not put such claims of a differential increase in the text. On the contrary, we rather highlighted that SPP1 can be induced by other non-neurodegenerative clinical conditions represented in the controls (see the discussion).

B.94 is indeed representing NEFL data but from the plasma (not CSF) and from the Bead array (not ELISA). We included this in the first submission but it was dropped during the review process.

Still, I understand that the current data and descriptions we put on GitHub can be misunderstood and you were very much right in contacting us.

We will soon update the GitHub repo with clearer descriptions of beads IDs and use of the analytics code.

Thank you again for your interest in our work and for reaching out.

With kind regards

Sebastian

Sebastian Lewandowski PhD

Karolinska Institutet, CMM, L8:04 Karolinska vägen 8A SE-171 64 Solna Twitter @LewandowskiSA

From: Andrew Jaffe @.***> Sent: Friday, April 16, 2021 7:40 PM To: lewandowskilab/PVF_Manuscript Cc: Sebastian Lewandowski; Comment Subject: Re: [lewandowskilab/PVF_Manuscript] different directionality in Figure 3? (#1)

Looking at this a bit more, I wonder if the README should state that RawB.94 is actually SPP1?

From README: RawB.53, RawB.123 and RawB.94 These are the log normalized MFI values where;

Distributions of each variable by Dx

@.***D732C5.F7B8ABA0]

Corresponding Figure 3a: @.***D732C5.F7B8ABA0]

Thanks!

From: lewandowskilab @.> Sent: Friday, April 16, 2021 5:04 AM To: lewandowskilab/PVF_Manuscript @.> Cc: Andrew Jaffe @.>; Author @.> Subject: Re: [lewandowskilab/PVF_Manuscript] different directionality in Figure 3? (#1)

Hello Andrew, Thanks for your insight. I will look into the matter and get back to you. Best wishes

Sebastian

Sebastian Lewandowski PhD

Karolinska Institutet, CMM, L8:04 Karolinska vägen 8A SE-171 64 Solna Twitter @LewandowskiSA

From: Andrew Jaffe @.<mailto:@.>> Sent: Friday, April 16, 2021 5:47 AM To: lewandowskilab/PVF_Manuscript Cc: Subscribed Subject: [lewandowskilab/PVF_Manuscript] different directionality in Figure 3? (#1)

Super impressive paper and efforts, and I applaud your group for making the data underlying your clinical associations available. However, it seemed like the directionality of SPP1 was the opposite in the Rdata object compared to Figure 3a - is there some other form of normalization necessary?

load("2020_02_Combined_Cohort_Dataframes copy.Rdata")

recode

cohorts$dx = cohorts$class cohorts$dx[cohorts$dx == "Neuro control"] = "Control" cohorts$dx[cohorts$dx == "ALS"] = "Case" cohorts$dx = droplevels(cohorts$dx) table(cohorts$dx)

Case Control

479 422

t.test(RawB.53 ~ dx, data= cohorts)

data: RawB.53 by dx

t = -4.6369, df = 673.65, p-value = 4.249e-06

alternative hypothesis: true difference in means is not equal to 0

95 percent confidence interval:

-0.1500042 -0.0607571

sample estimates:

mean in group Case mean in group Control

7.358855 7.464236

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Flewandowskilab%2FPVF_Manuscript%2Fissues%2F1&data=04%7C01%7Csebastian.lewandowski%40ki.se%7Cfd37dc44df0c44f31b6308d9008a6b8d%7Cbff7eef1cf4b4f32be3da1dda043c05d%7C0%7C0%7C637541416780869394%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=AAuph0C0ewWC9Ibs5Q7H%2FpiJCfXuy2qX9EGNgWkyNU0%3D&reserved=0, or unsubscribehttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAMHSJKB6WX7NRBA5677NVXTTI6XOVANCNFSM43AWVHMQ&data=04%7C01%7Csebastian.lewandowski%40ki.se%7Cfd37dc44df0c44f31b6308d9008a6b8d%7Cbff7eef1cf4b4f32be3da1dda043c05d%7C0%7C0%7C637541416780869394%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=NNq%2F%2B1Is9vPfGdiIa9yPtVtVTew%2BuFF4qkDqbgfC21U%3D&reserved=0.

När du skickar e-post till Karolinska Institutet (KI) innebär detta att KI kommer att behandla dina personuppgifter. Här finns information om hur KI behandlar personuppgifterhttps://ki.se/medarbetare/integritetsskyddspolicy.

Sending email to Karolinska Institutet (KI) will result in KI processing your personal data. You can read more about KI’s processing of personal data herehttps://ki.se/en/staff/data-protection-policy.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/lewandowskilab/PVF_Manuscript/issues/1#issuecomment-821033571, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ARC3BQRJERLCRPGM6JUHAVTTI74PZANCNFSM43AWVHMQ.

— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Flewandowskilab%2FPVF_Manuscript%2Fissues%2F1%23issuecomment-821332622&data=04%7C01%7Csebastian.lewandowski%40ki.se%7C1864c376837342ed954c08d900feb0a6%7Cbff7eef1cf4b4f32be3da1dda043c05d%7C0%7C0%7C637541916147932811%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=8ZtHSm3RZaCwGV4DVpbgDR4%2Bqyj%2FldQ7Scm2HD9WzBI%3D&reserved=0, or unsubscribehttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAMHSJKCGCSE5BRZJK3PARNTTJBY7ZANCNFSM43AWVHMQ&data=04%7C01%7Csebastian.lewandowski%40ki.se%7C1864c376837342ed954c08d900feb0a6%7Cbff7eef1cf4b4f32be3da1dda043c05d%7C0%7C0%7C637541916147932811%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=9nLWdLaKRR4TMhH2wAXXKSqw6sKY%2BEi0YFbyRkSPlsY%3D&reserved=0.