Issues Using DMRcate and rmSNPandCH

Question

[Indira]

Technical Specs:

• Working on an HPC that is Linux
• My machine is a Mac Air
• I am using an R Markdown
• My R version is 4.2.2

I am using the following libraries:

library("limma")
library("minfi")
library("RColorBrewer")
library("missMethyl")
library("Gviz")
library("DMRcate")
library("stringr")
library("mCSEA")
library("IlluminaHumanMethylationEPICmanifest")

My basic code used before I encountered the code with an error is below

targets <- read.metharray.sheet(Directory, pattern = "My_Targets_File.csv")
RGset <- read.metharray.exp(targets = targets)
detP <- detectionP(RGset)
head(detP)
GRset <- preprocessFunnorm(RGset)
m <- getM(GRset)

The DMRcate manual (https://bioconductor.org/packages/devel/bioc/vignettes/DMRcate/inst/doc/DMRcate.pdf) suggests that probes that are close to SNPs should be filtered out.

My code that generated a rather annoyingly reoccurring error is below:

m.noSNPs <- rmSNPandCH(m, dist=2, mafcut=0.05)

It give the following error:

Cannot connect to ExperimentHub server, using 'localHub=TRUE' instead

Using 'localHub=TRUE' If offline, please also see BiocManager vignette section on offline use

snapshotDate(): 2023-01-20

Error: File not previously downloaded. Run with 'localHub=FALSE' Traceback:

1. rmSNPandCH(m, dist = 2, mafcut = 0.05)
2. eh[["EH3130"]]
3. eh[["EH3130"]]
4. .local(x, i, j = j, ...)
5. stop("File not previously downloaded.\n", " Run with 'localHub=FALSE'", . call. = FALSE)

I do not need the ExperimentalHub as I have my own data from EPIC 2.0. What can I do to navigate around this error/correct my code? And why is the snapshot date Jan 20,2023 - its Feb 4th?

When I try to use the localHub = TRUE argument this happens:

m.noSNPs <- rmSNPandCH(m, dist=2, mafcut=0.05, localHub= TRUE)

It gives me the error:

Error in rmSNPandCH(m, dist = 2, mafcut = 0.05, localHub = TRUE): unused argument (localHub = TRUE) Traceback:

If I try LocalHub = FALSE:

m.noSNPs <- rmSNPandCH(m, dist=2, mafcut=0.05, localHub= FALSE)

I get the following error:

Error in rmSNPandCH(m, dist = 2, mafcut = 0.05, localHub = FALSE): unused argument (localHub = FALSE) Traceback

I am using R - 4.2.2 and the latest version of DMRcate. I have reinstalled DMRcate and specifically rmSNPandCH.

Then I added the library for ExperimentHub to the list of libraries:

library("ExperimentHub")

However then when I ran the command:

m.noSNPs <- rmSNPandCH(m, dist=2, mafcut=0.05)

I still get the following error:

Cannot connect to ExperimentHub server, using 'localHub=TRUE' instead

Using 'localHub=TRUE' If offline, please also see BiocManager vignette section on offline use

snapshotDate(): 2023-01-20

Error: File not previously downloaded. Run with 'localHub=FALSE' Traceback:

1. rmSNPandCH(m, dist = 2, mafcut = 0.05)
2. eh[["EH3130"]]
3. eh[["EH3130"]]
4. .local(x, i, j = j, ...)
5. stop("File not previously downloaded.\n", " Run with 'localHub=FALSE'", . call. = FALSE)

Then I tried with localHub = TRUE:

m.noSNPs <- rmSNPandCH(m, dist=2, mafcut=0.05, localHub=TRUE)

However, I got the following error:

Error in rmSNPandCH(m, dist = 2, mafcut = 0.05, localHub = TRUE): unused argument (localHub = TRUE) Traceback:

Then, I tried with localHub = FALSE:

m.noSNPs <- rmSNPandCH(m, dist=2, mafcut=0.05, localHub=FALSE)

Sadly, I then got the error below:

Error in rmSNPandCH(m, dist = 2, mafcut = 0.05, localHub = FALSE): unused argument (localHub = FALSE)

Traceback:

I also tried loading the library for DMRcate data in case that would help however, that didn't seem to stop the same error that has been plaguing me:

Code used:

library("DMRcatedata")

Error:

Cannot connect to ExperimentHub server, using 'localHub=TRUE' instead

Using 'localHub=TRUE' If offline, please also see BiocManager vignette section on offline use

snapshotDate(): 2023-01-20

Error: File not previously downloaded. Run with 'localHub=FALSE' Traceback:

1. rmSNPandCH(m, dist = 2, mafcut = 0.05)
2. eh[["EH3130"]]
3. eh[["EH3130"]]
4. .local(x, i, j = j, ...)
5. stop("File not previously downloaded.\n", " Run with 'localHub=FALSE'", . call. = FALSE)

Forcing a re-install did not make the error go away:

BiocManager::install("DMRcate", force = TRUE)

---

Proxy Related Code (try):

I also tried the proxy related commands detailed in this link: https://support.bioconductor.org/p/132864/

However, it did not work.

Attempt 1:

library(ExperimentHub)
library(httr)
setExperimentHubOption("PROXY", "my_proxy")
proxy = getExperimentHubOption("PROXY")
GET("https://experimenthub.bioconductor.org/metadata/experimenthub.sqlite3", proxy="my_proxy")

Attempt 1 Error:

Error in curl::curl_fetch_memory(url, handle = handle): Timeout was reached: [experimenthub.bioconductor.org] Connection timed out after 10000 milliseconds Traceback:

1. GET("https://experimenthub.bioconductor.org/metadata/experimenthub.sqlite3", . proxy = "my_proxy")
2. request_perform(req, hu$handle$handle)
3. request_fetch(req$output, req$url, handle)
4. request_fetch.write_memory(req$output, req$url, handle)
5. curl::curl_fetch_memory(url, handle = handle)

Attempt 2:

GET("https://experimenthub.bioconductor.org/metadata/experimenthub.sqlite3")

Attempt 2 Error:

Error in curl::curl_fetch_memory(url, handle = handle): Timeout was reached: [experimenthub.bioconductor.org] Connection timed out after 10001 milliseconds Traceback:

1. GET("https://experimenthub.bioconductor.org/metadata/experimenthub.sqlite3")
2. request_perform(req, hu$handle$handle)
3. request_fetch(req$output, req$url, handle)
4. request_fetch.write_memory(req$output, req$url, handle)
5. curl::curl_fetch_memory(url, handle = handle)

Attempt 3:

setExperimentHubOption("PROXY","my_proxy")
Sys.setenv(EXPERIMENT_HUB_PROXY = Sys.getenv("my_proxy"))

Then I ran:

m.noSNPs <- rmSNPandCH(m, dist=2, mafcut=0.05)

Attempt 3 Error:

Assuming valid proxy connection through 'my_proxy' If you experience connection issues consider using 'localHub=TRUE'

Warning message: "Could not check database for updates. Database source currently unreachable. This should only be a temporary interruption. Using previously cached version." Error in value[3L]: failed to connect reason: Failed to connect to bioconductor.org port 443: Connection timed out Consider rerunning with 'localHub=TRUE' Traceback:

1. rmSNPandCH(m, dist = 2, mafcut = 0.05)
2. ExperimentHub()
3. .Hub("ExperimentHub", hub, cache, proxy, localHub, ask, ...)
4. tryCatch({ . db_date <- .restrictDateByVersion(db_path) . }, error = function(err) { . stop("failed to connect", "\n reason: ", conditionMessage(err), . "\n Consider rerunning with 'localHub=TRUE'") . })
5. tryCatchList(expr, classes, parentenv, handlers)
6. tryCatchOne(expr, names, parentenv, handlers[[1L]])
7. value[3L]
8. stop("failed to connect", "\n reason: ", conditionMessage(err), . "\n Consider rerunning with 'localHub=TRUE'")

Attempt 4:

This code was used in R in Linux command line and it worked however, when I chose to run the commands I needed in R Markdown I still could not get rmSNPandCH to work:

library(ExperimentHub)
library(httr)
setExperimentHubOption("PROXY", "my_proxy")
proxy = getExperimentHubOption("PROXY")
GET("https://experimenthub.bioconductor.org/metadata/experimenthub.sqlite3", proxy="my_proxy")

The following code was run in R Markdown:

m.noSNPs <- rmSNPandCH(m, dist=2, mafcut=0.05)

I got the following error:

Assuming valid proxy connection through 'my_proxy' If you experience connection issues consider using 'localHub=TRUE'

Warning message: "Could not check database for updates. Database source currently unreachable. This should only be a temporary interruption. Using previously cached version." Error in value[3L]: failed to connect reason: Failed to connect to bioconductor.org port 443: Connection timed out Consider rerunning with 'localHub=TRUE' Traceback:

1. rmSNPandCH(m, dist = 2, mafcut = 0.05)
2. ExperimentHub()
3. .Hub("ExperimentHub", hub, cache, proxy, localHub, ask, ...)
4. tryCatch({ . db_date <- .restrictDateByVersion(db_path) . }, error = function(err) { . stop("failed to connect", "\n reason: ", conditionMessage(err), . "\n Consider rerunning with 'localHub=TRUE'") . })
5. tryCatchList(expr, classes, parentenv, handlers)
6. tryCatchOne(expr, names, parentenv, handlers[[1L]])
7. value[3L]
8. stop("failed to connect", "\n reason: ", conditionMessage(err), . "\n Consider rerunning with 'localHub=TRUE'")

Attempt 5:

I then went to the command line in Linux and ran the following:

wget https://experimenthub.bioconductor.org/metadata/experimenthub.sqlite3

Attempt 5 Error:

Then when I tried to run the m.noSNPs <- rmSNPandCH(m, dist=2, mafcut=0.05) command I got the following error:

Assuming valid proxy connection through 'my_proxy' If you experience connection issues consider using 'localHub=TRUE'

Warning message: "Could not check database for updates. Database source currently unreachable. This should only be a temporary interruption. Using previously cached version." Error in value[3L]: failed to connect reason: Failed to connect to bioconductor.org port 443: Connection timed out Consider rerunning with 'localHub=TRUE' Traceback:

1. rmSNPandCH(m, dist = 2, mafcut = 0.05)
2. ExperimentHub()
3. .Hub("ExperimentHub", hub, cache, proxy, localHub, ask, ...)
4. tryCatch({ . db_date <- .restrictDateByVersion(db_path) . }, error = function(err) { . stop("failed to connect", "\n reason: ", conditionMessage(err), . "\n Consider rerunning with 'localHub=TRUE'") . })
5. tryCatchList(expr, classes, parentenv, handlers)
6. tryCatchOne(expr, names, parentenv, handlers[[1L]])
7. value[3L]
8. stop("failed to connect", "\n reason: ", conditionMessage(err), . "\n Consider rerunning with 'localHub=TRUE'")

Attempt 6:

I ran the following in command line:

export my_proxy_1
export my_proxy_2

Attempt 6 Error:

Then restarted my Markdown from scratch to see if it would work and for the rmSNPandCH I used I got the following error:

Cannot connect to ExperimentHub server, using 'localHub=TRUE' instead

Using 'localHub=TRUE' If offline, please also see BiocManager vignette section on offline use

snapshotDate(): 2023-01-20

Error: File not previously downloaded. Run with 'localHub=FALSE' Traceback:

1. rmSNPandCH(m, dist = 2, mafcut = 0.05)
2. eh[["EH3130"]]
3. eh[["EH3130"]]
4. .local(x, i, j = j, ...)
5. stop("File not previously downloaded.\n", " Run with 'localHub=FALSE'", . call. = FALSE)

• I also tried to see if I could use this Troubleshoot The Hubs manual (https://bioconductor.org/packages/devel/bioc/vignettes/AnnotationHub/inst/doc/TroubleshootingTheCache.html#group-hubcache-access) - however, I am not sure how I can use it in this case for this issue

I am trying to use the rmSNPandCH command for my own data can someone tell me how to navigate around these errors? Can you give me sample code to do so?

Answer 1

[t.peters]

Hi Indira,

Apologies for the delay in getting to your post. For the moment, DMRcate does not have compatibility with EPICv2. However, I anticipate it will be updated in the coming months. I will update the ExperimentHub() data object with SNP information from the new manifest, as well as cross-hybridising probes, and update rmSNPandCH() to include this functionality. Thank you for your patience.

As for the ExperimentHub() error, I think this is a separate connectivity issue. Try posting your EH issues directly to the Bioconductor support page (https://www.bioconductor.org/help/support/) and I imagine someone will be able to help you with this.

Cheers, Tim

Answer 2

[t.peters]

Hi Indira,

DMRcate can now call DMRs from EPICv2 in the current devel version: https://www.bioconductor.org/packages/devel/bioc/html/DMRcate.html. You can find the vignette for EPICv2 assays here: https://www.bioconductor.org/packages/devel/bioc/vignettes/DMRcate/inst/doc/EPICv2.pdf

This version will become 3.0 in the next Bioconductor release, available May 1st.

If you're interested, you can also check out our latest paper in BMC Genomics (https://bmcgenomics.biomedcentral.com/articles/10.1186/s12864-024-10027-5), where we take a deep dive into many aspects of EPICv2, including reproducibility and multiple probes mapping to the same CpG site.

Cheers, Tim