Showing posts with label statistics. Show all posts
Showing posts with label statistics. Show all posts

Wednesday, 19 September 2012

I used to think correlation implied causation ...



Alerted to this by 

@mbeisen

Biologist at UC Berkeley & HHMI; open access advocate and co-founder of Public Library of Science
Berkeley, CA · http://www.michaeleisen.org


Thursday, 17 March 2011

Common numbers / statistics for Uniquely mapped reads?

Was asked if there was a commonly reported numbers for
uniquely mapped reads (which is troublesome to define with bowtie)
vs
total mapped reads

Not sure also if the numbers differ for applications
e.g.
WGS
exome reseq

human vs other organisms.
Got this figure from a 2009 paper. Not sure if anyone collates data like this
http://bioinformatics.oxfordjournals.org/content/25/7/969.full.pdf

Saturday, 12 March 2011

RStudio IDE for R ... looks promising!

RStudio, released yesterday, is a new open-source IDE for R. It’s getting a lot of attention at R-bloggers and it’s easy to see why: this is open-source software development done right.
from 

What You’re Doing Is Rather Desperate

Notes from the life of a bioinformatics researcher



Friday, 9 April 2010

bowtie build time statistics / benchmark

Recently Built colorspace index for hg19 (from ftp://ftp.sanger.ac.uk/pub/1000genomes/tk2/main_project_reference/)

AMD Phenom II X4 955 chip
8 GB RAM

SATA2 HDD

Wrote 822714402 bytes to primary EBWT file: hg19.rev.1.ebwt
Wrote 358098108 bytes to secondary EBWT file: hg19.rev.2.ebwt
Re-opening _in1 and _in2 as input streams
Returning from Ebwt constructor
Headers:
    len: 2864784823
    bwtLen: 2864784824
    sz: 716196206
    bwtSz: 716196206
    lineRate: 6
    linesPerSide: 1
    offRate: 5
    offMask: 0xffffffe0
    isaRate: -1
    isaMask: 0xffffffff
    ftabChars: 10
    eftabLen: 20
    eftabSz: 80
    ftabLen: 1048577
    ftabSz: 4194308
    offsLen: 89524526
    offsSz: 358098104
    isaLen: 0
    isaSz: 0
    lineSz: 64
    sideSz: 64
    sideBwtSz: 56
    sideBwtLen: 224
    numSidePairs: 6394609
    numSides: 12789218
    numLines: 12789218
    ebwtTotLen: 818509952
    ebwtTotSz: 818509952
Total time for backward call to driver() for mirror index: 01:24:19

Gosh and I thought it will take hours!

Datanami, Woe be me