Data for biomolecular sciences

I’m not sure whether presentations are more like curries or sponge cakes. A curry is usually better if you prepare it in advance and leave it to mature for a day or two. A sponge is nicest when it is fresh. There are of course limits. If you leave the curry for too long there is a danger that several days later you will find it mouldering in a corner of the fridge, quite inedible. On the other hand, if you try to eat a sponge just as it has left the oven it will tend to fall apart.
Well, I have one week left to prepare a talk on “Data for biomolecular sciences” and I really don’t want to fall apart. I’ve given the talk a subtitle “a quick skirmish” as it is intended as a high-level (i.e. superficial) overview. I think it is impossible to do more than that in just 50 minutes. The audience is fellow librarians, people wanting to support information-seekers unfamiliar with these tools. I have done similar talks before but not the last was in 2003 and I am a bit rusty. Back then I borrowed a fair bit from the User’s guide to the human genome published in Nature Genetics around that time. Sadly this has not been updated since September 2003 and I’m not sure how good a guide it is now. The excellent Current Protocols in Bioinformatics helps, as do training materials at NCBI and EBI.
A number of sites of course demand to be mentioned – EBI, NCBI, Uniprot, OMIM, PDB and possibly KEGG. Trying to choose what to show on the bigger sites is hard. I’ve heard praise for other sites such as WikiGenes, WikiProteins and IHOP. Jenny Rohn’s post on gene names gave some useful clues to practical utility.
I’d be grateful for more guidance from your good selves:

1. What are your essential everyday resources for biomolecular data?
2. What are the typical questions you need to answer (and is the 2003 Nature Genetics guide still useful)?

1. Can you recommend any more tutorials / guides?

Thanks!