Saturday, 15 November 2014

What 5G mobile networks portends for the future of personal genomics

ok I saw this a while back (a month ago, yes I have been busy)

I am already very impressed with 4G (LTE) speeds but with 5G you can possibly achieve 150 mb/s to 940 mb/s which is mind blowing ...

Considering that you could then possibly upload via your mobile devices, your own 100 Gb bam file in about 10 seconds (sorry I wasn't thinking how much faster a youtube video would stream). Now Google is saying that they can store your genome (actually they meant your 30x WGS bam file) for $25 a year. But with 5G speeds, why would I even bother with that?

Heck, maybe in the future with an USB OTG cable connected to Oxford Nanopore's MinION your android phone will be able to sequence and upload in realtime your DNA obtained from a buccal swab. The cloud will have the fastq reads aligned and call variants instantaneously and download the 100 Gb bam to your microsd card.

Possible applications:

  1. Maybe in the future other than asking if you have a drug allergy, pharmacists will request to 'scan' your DNA for the most efficient drug. 
  2. another possible application might be having your DNA be your own personal identity card, 
  3. more routine sequencing of the human microbiome to monitor your health in relation to the gut microflora or other sites.

I am keen to find out what you think you can do if you could carry your whole genome sequencing with you and upload via mobile networks. Drop in your comments please!

Saturday, 13 September 2014

Finally getting to see my ubiome data ....

Wonder if I can access the raw data though ...

Wednesday, 27 August 2014

tabix and VCF file size limits

Today I learn that tabix can index bgzipped VCF files of 4 TB (compressed) and possibly bigger.... Mind blown ...

Source: Samtools-help mailing list

Friday, 2 May 2014

Monday, 21 April 2014

Fwd: Welcome to the Google Genomics Preview


---------- Forwarded message ----------

Welcome to the Google Genomics Preview! You've been approved for early access to the API.

The goal of the Genomics API is to encourage interoperability and build a foundation to store, process, search, analyze and share tens of petabytes of genomic data.

We've loaded sample data from public BAM files:

* The complete 1000 Genomes Project

* Selections from the Personal Genome Project

How to get started:

* Follow the instructions in the developer documentation

* Try the sample genome browser which calls the API

* Try out the other open source examples -- an R script, Python MapReduce, and a Java file-based implementation

* Write your own code to call the API and explore new uses

This is only the beginning. Your feedback will be essential to make the API useful. Please submit feature requests, bugs and suggestions on our GitHub page.

Thank you for being part of the first wave. If you'd rather join with a different email address (Gmail or Google Apps domain), please fill out the request form with that address too, and we'll grant access soon. Thank you for your interest!


The Google Genomics team

Tuesday, 25 March 2014

Monocle is a R toolkit for analyzing single cell expression experiments

---------- Forwarded message ----------
From: Cole Trapnell <cole>
Date: Mon, Mar 24, 2014 at 11:51 PM
Subject: [Bowtie-bio-announce] Monocle 0.99.0
To: bowtie-bio-announce

I am proud to announce the first release of the Monocle analysis toolkit for single-cell RNA-Seq and qPCR.  Monocle performs differential expression and time series analysis for single-cell expression experiments.  You can read about Monocle and its “pseudo time” analysis of biological processes in the paper, which just appeared on Nature Biotech’s AOP list:

The Monocle source code and support site is also live:

I hope you will consider using Monocle for your single-cell expression analysis workflow!  Please report any issues on the new Monocle google group:!forum/monocle-users

Note that Monocle is currently considered an ALPHA release - new features and interfaces changes will be coming in future releases, along with significant new functionality. The release announcement for Monocle v 0.99.0 is below.



0.99.0 release - 3/23/2014

The first public release of Monocle is now available for download. Monocle is a toolkit for analyzing single cell expression experiments. It runs on the R statistical computing platform.

Monocle takes as input a matrix of gene or transcript expression values, along with information about each cell in the experiment, and some annotation about each gene/transcript, both provided as simple tables. It is designed for single-cell RNA-Seq experiments, but in principle Monocle can be used with other data types such as qPCR. For RNA-Seq, you can useCufflinks to estimate your expression values.

This software is a work in progress - it is a beta release, and new features will continue to be added over the next couple of weeks. To suggest a feature or report a bug, please post your comments on the Monocle user's group.

Cole Trapnell

Wednesday, 5 March 2014

Google Genomics -- Google Developers Google genomics limited preview is out.
Been wondering when and how they might do genomics.
Now suddenly it's scarier to link all ur google accounts together. 

Maybe we will see gene association studies with different online surfing habits very soon! *chuckles*

Thursday, 27 February 2014

It's been a while ... Python 3 print is now a function

Gosh it's definitely telling that I haven't been coding in Python 3 for a while.

I didn't know that they have changed the print statement to a function. So now I need parentheses

for details see

Tuesday, 28 January 2014

Growth slows in cloud business

Has the cloud hype died down? Seagate reportedly misses analyst's estimates as growth slowed in it's cloud storage business. With the exception of Galaxy or Basespace, it's hard to get researchers to buy into the cloud analysis paradigm even though their sporadic usage pattern would fit nicely for the economics of cloud usage. Maybe the cost of storage space / moving files into the cloud is the biggest hurdle.  My two cents.

Datanami, Woe be me