sirselim/illumina450k_filtering

A collection of resources to filter 'bad' probes from the Illumina 450k and EPIC methylation arrays

---

---

👋 Hi there! I'm Miles

I'm a bioinformatician / computational geneticist based in New Zealand. I am the Associate Director of the APAC Applications Project Support team at Oxford Nanopore Technologies. I also dabble at photography and guitar, with a side of coffee, beer and single malts in the mix.

I have written and maintiain several documents about GPU compute for Nanopore sequencing that people find helpful, feel free to check them out:

• GPU musings (with an eye on genomics)
• GPU price / performance comparisons for Nanopore basecalling
• Nanopore Guppy GPU basecalling on Windows using WSL2
• Nvidia Jetson Xavier Community Gist
• Nvidia Jetson Nanopore Sequencing - GitHub Repository

You can find out more about me and connect in different ways below:

---

---

---

---

illumina450k_filtering's Issues

Run before or after normalisation?

Hello, you mention in the README that these filtering steps must be run during the QC stage of an analysis. Yet in the code the beta value object used is named beta_norm. Should the beta values used be computed before or after normalisation?

Using minfi:

Exemple code for normalization:
mSetSq <- preprocessQuantile(RGSet)

Exemple code to get beta values after normalisation:
beta_norm <- getBeta(mSetSq)

Exemple code to get beta values after normalisation:
beta_norm <- getBeta(RGSet)

Thank you.

This is great. How do we cite this?

I would like to cite this. Is there a reference?

citations?

Hi, thank you for aggregating this. Is there a way to cite your work other than this github?

final list of EPIC cross reactive probes seems high

I got 135885 probes in the final list of unique probes, wondering if it is normal. This is high % of probes that I suppose to removed.

Other locations of BOWTIE2 mapping of 450k probes

Dear sirselim,
Do you have any additionnal file listing locations where the multiple alignments happened for the BOWTIE2 mapping of 450k probes ?
Something in Bed-like format:

probeID | chr | start | end
cg00000658 | 1 | 15454 | 15500
cg00000658 | 7 | 457851 | 457900
...

I would be very interested in it if you could add it.
Thank you for making available this repository.

Best regards.

Other Chen et al probes

Hello @sirselim ,

Thank you for hosting the lists of reactive probes from Chen et al 2013 paper in this repo. The host website seems to have been redesigned recently, and they are nowhere to be found there.

I was curious though if by any chance you also curate their list of polymorphic probes- or know of such a mirror?

Best,
Azza