RNA Methylation

Investigations into the prevalence and role of RNA methylationseem to be gathering pace, and receiving attention via high impact publications. I think this is an exciting new level of regulation ofRNAs, and could be key to understanding processes such as translational control and localization.

Two papers in particular, Dominissini et al, 2012, Nature [DOI] and Wang et al, 2014, Nature [DOI] have caught my attention. They both investigate the role of N6–methyladenosine, the most common non-cap mRNA modification.

Dominissini et al, produced an RNAmethylome (m6A) map for human and mouse using m6A–Seq (antibody pulldown followed by high throughput sequencing). Wang et al, demonstrated m6A-dependent regulation of mRNA stability, through recoginition ofm6A by YTDHDF2. Each paper defines an m6A consensus motifs:Gm6AC (Wang et al, Nature 2014) and RRm6ACH (or as a regular expression [AG][AG]AC[ACU]) (Dominissini et al, 2012, Nature).

I think these studies have only just begun to scratch the surface of the roles m6A methylation can play in translational regulation, and I look forward to reading more studies into the importance ofm6A.

Posted in RNA

Misunderestimating the significance of p-values

In (biological) science there is an expectation that any quantitative evidence should be accompanied with a statistical significance metric such as a p-value or error bars. These values are then deemed to be significant if they fall within certain confidence intervals. e.g. 99.9% confidence interval for a p-value of 0.01. I personally believe p-values should only be considered significant at extremely high confidence intervals with 99.9% a bare minimum, ideally 99.999% or higher. But this obviously depends on the experiment and sample size.

Recently there have been two excellent papers addressing the issue of p-value misuse. One late last year by Regina Nuzzo in PNAS [DOI] and the other this week by Valen Johnson in Nature [DOI]. Both paper discuss the affect the p-value has on data (and conclusion) reproducibility.

This is a difficult issue to address as the current p-value usage is so ingrained in the scientific community. I see the only effective way of addressing this is through journals requiring p-values be used correctly accompanied with all the data provided open access.

To preprint of not to preprint

There has been a lot of discussion around the openness of scientific publishing. Papers should be freely available to everyone upon publication so they reach the largest possible audience, disseminating science often funded by charitable bodies. But, as I am in the process of putting together my next paper, I am weighing up the pros and cons of uploading my finished manuscript to a preprint server bioRxiv before submitting to my journal of choice.

Pros: My paper gets out into the world and is citable immediately and I get feedback from comments to improve the paper (see cons below)

Cons: Preprint servers for biological papers are pretty new, so there is a developing community for commenting on the paper. So one of the main pros for preprinting, might not be currently viable. i.e. I don’t get any comments. I read a lot of papers on bioRxiv and very few have attracted comments so far. Not all journals accept papers previously submitted to a preprint server. This could reduce my choice of journals to submit to (although a lot of my journals of choice do accept preprinted papers, and more journals are updating their policies)

Conclusion: I like the principal of using preprint servers, and I will be using bioRxiv for my next paper. However, I may not immediately benefit from doing so. By using bioRxiv now, I hope to do my bit for open publishing and contribute to the bioRxiv community. In particular I think it’s important to contribute comments to the papers I read. What would be good to see is scientists being invited by bioRxiv to add comments to a paper.

Laptops for RNA Bioinformatics

As my current laptop, MacBookPro 15″ is approaching retirement I have been looking at possible replacements. I use my laptop for programming / data analysis, but mostly paper writing, paper figure making, and presentations. I have a high-spec Linux workstation and access to Linux CPU/storage nodes, which take care of all my processing (NGS data, RNA/protein structure prediction and RNA motif searching). Although I have an Apple at the moment, I’m not a Mac fan-boy.

My ideal bioinformatics laptop should have:

Software: I use different combinations of document preparation software to align with collaborators, I would probably just use LaTeX/BibTeX if given the choice. I use lots of web/cloud based tools, e.g. Evernote, Google Play, but these should work on any platform.

Programming (Perl, Python, R, C all via Vi with Git etc)
A terminal for SSH and Vi
MS Word (with Papers/Endnote)
LaTex (Texmaker with Git)
Powerpoint (and possibly keynote)
Adobe Illustrator and Inkscape (or just Inkscape at a stretch)
Hardware: My laptop goes everywhere with me including my commute by bike. In the office I hook it up to a large external monitor, but I still need a decent laptop screen (matt, high res) for at home and travelling.

Minimum of 512Gb SSD
Minimum of 8Gb RAM, ideally 16GB
Intel Haswell Chips (i7)
Battery Life, I’m rarely more than 5 hours from a power supply.
Matt screen (glossy screens reflect office lights too much). Unlikely to be an option as I can’t find any laptops with this option.
Light weight. This also includes the power-brick
13″ Screen size (11″ is a little small for programming, 15″ too big for travelling)
Current options: So what is available on the market to (nearly) meet these specifications? There is the option of dual booting to allow OSX/Windows with Linux or a virtual machine (Linux or Windows). Neither of which are ideal.

Apple MacBookAir 13″ [Apple]
Apple MacBookPro 13″ retina display [Apple]
Dell XPS 13 Developer Edition [Dell]
Razer Blade High performance Ultrabook [razer-blade]
A wide selection of Windows 8 Ultrabooks
Conclusion: My ideal laptop is the Razer Blade, but with a matt screen, running Linux, native Word/Powerpoint etc and available in the UK. However this doesn’t exist… yet. So I’m left with either another Apple (probably an Air) or a Dell running a Windows virtual machine. Neither option does everything I want. I hope Dell will continue the Developer Edition laptops but with higher spec hardware (bigger SSD and better screen in particular).