Missing Human lncRNA by design?

[AndrewSkelton]

No description provided.

[AndrewSkelton]

Hi,

I've been looking at the hg38 assets, and taking noticed that a lncRNA (MALAT1, ENST00000619449) appears to be missing from annotation (Salmon Index + tx2gene map). Is this by design? - ie is it mostly protein coding genes included in the index?

Additionally, there used to be a full decoy Salmon index for hg38 in the bundle but I can't find that anymore, has it been deprecated?

Thanks (apologies if I've missed an obvious answer somewhere!)

Andrew

[stolarczyk]

hi @AndrewSkelton, thanks for reaching out.

Which assets are you exactly referring to? These ones?

• hg38/salmon_partial_sa_index
• hg38/tgMap

We definitely didn't remove any sequences/transcripts on purpose. The assets are built following recipes (which you can find on the pages linked above) that were established by the salmon authors. The sequence may be missing in the reference sequence, that was used to create the assets.

We're willing to accommodate any changes as long as they are required for the general public.

Keep in mind that you can build the assets in question locally, using any reference data you wish: http://refgenie.databio.org/en/latest/build/