Latex scientific notation, made easy

\providecommand{\e}[1]{\ensuremath{\times 10^{#1}}}

Then, typing

The [111] crystal planes are 3.2\e{-10} m apart.

http://www.tapdancinggoats.com/easy-scientific-notation-in-latex.htm

R plot magic

To move the axis labels/lines:

par(mpg=c(, , ))

default is mpg=c(3,1,0), but mgp=c(1.75, 0.5, 0) works

To move margins:

par(mar=c(,,,))

default is mar=c(5,4,4,2)+0.1, but mar=c(3,3,0,0)+0.1 works

Using awk to randomly sample a file

Create a file with 1000 lines:

for i in {1..1000};do echo $i; done > f


export P=0.5
cat f | awk -v p=$P 'BEGIN{srand()} {r = rand(); if (r <= p) print}'


Best Latex Web site EVER

http://detexify.kirelabs.org/classify.html

General Papers

Next-generation gap

http://www.nature.com/nmeth/journal/v6/n11s/full/nmeth.f.268.html

 

Downloading an Entire Web Site with wget

$ wget \
     --recursive \
     --no-clobber \
     --page-requisites \
     --html-extension \
     --convert-links \
     --restrict-file-names=windows \
     --domains website.org \
     --no-parent \
         www.website.org/tutorials/html/

This command downloads the Web site www.website.org/tutorials/html/.

The options are:

• –recursive: download the entire Web site.
• –domains website.org: don’t follow links outside website.org.
• –no-parent: don’t follow links outside the directory tutorials/html/.
• –page-requisites: get all the elements that compose the page (images, CSS and so on).
• –html-extension: save files with the .html extension.
• –convert-links: convert links so that they work locally, off-line.
• –restrict-file-names=windows: modify filenames so that they will work in Windows as well.
• –no-clobber: don’t overwrite any existing files (used in case the download is interrupted and
  resumed).

http://www.linuxjournal.com/content/downloading-entire-web-site-wget

My 5c

http://my5c.umassmed.edu/welcome/welcome.php

Local Installation and Use of R Packages

http://csg.sph.umich.edu/docs/R/localpackages.html

1. Specifying a local library search location

Specify a local library search location.

You can use several library trees of add-on packages. The easiest way to tell R to use these via a ‘dotfile’ by creating the following file ‘$HOME/.Renviron’ (watch the quotes and ~ character):

  R_LIBS_USER="~/R/library"

This specifies a keyword (R_LIBS_USER) which points to a colon-separated list of directories at which R library trees are rooted. You do not have to specify the default tree for R packages.

If necessary, create a place for your R libraries

  mkdir ~/R ~/R/library         # Only need do this once

Set your R library path

  echo 'R_LIBS_USER="~/R/library"' >  $HOME/.Renviron

2. Installing to a local library search location

Installation is dead easy. Start up R and tell R to fetch your package from CRAN, compile whatever needs compiling and set everything else up.

Beware – each package will only work for the platform (i.e. Linux or Solaris) where you installed it. If you want a package on both Linux and Solaris, you’ll need to install it in different directories for each system type.

  R                     # Invoke R
  > install.packages("name-of-your-package",lib="~/R/library")

Novoalign Alignment Scores

Base Qualities and Alignment Scores

Novoalign aligns reads against a reference genome using qualities and ambiguous nucleotide codes.
The initial alignment process finds alignment locations in the indexed sequence that are possible
sources of the read sequence. The alignment locations are scored using the Needleman­Wunsch
algorithm with affine gap penalties and with position specific scoring derived from the read base
qualities and any ambiguous codes in the reference sequence. User defined affine gap penalties are
used for scoring insert/deletes.
Novoalign uses Needleman­Wunsch alignments with affine gap penalties, the gap opening penalty
should be set to where is the probability of an insertion deletion
mutation vs the reference genome and is the gap extension penalty. Likewise the gap extend
penalty can be set to where is the probability of a single base indel
and is the probability of a 2 base insert/delete mutation. The default gap penalties were
derived from the frequency of short insert/deletes in human genome resequencing projects.
Base quality values are used to calculate base penalties for the Needleman ­Wunsch algorithm. The
base qualities are converted to base probabilities and then to score penalties.

PRB Quality to Score Conversion

The prb file has quality score for each base, , at each position, , in the read. The quality
value is converted to a probability, and then to a penalty .

Alignment Score and Threshold

The alignment score is where is the probability of the read sequence
given the alignment location .

A threshold of 75 would allow for alignment of reads with two mismatches at high quality base
positions plus one or two mismatches at low quality positions or to ambiguous characters in the
reference sequence.
If a threshold is not specified then Novoalign will calculate a threshold for each read such that an
alignment to a non­repetitive sequence will have an alignment quality of at least 20. I.e. The
iterative process of finding an alignment will terminate before finding a low quality chance
alignment. Alignments to repetitive sequences may still have qualities less than 20.

Posterior Alignment Probabilities and Quality Scores

The posterior alignment probability calculation includes all the alignments found; the probability
that the read came from a repeat masked region or from any regions coded in the reference genome
as N’s; and an allowance for a chance hit above the threshold based on the mutual information
content of the read and the genome.
A posterior alignment probability, is calculated as:

where is the probability of finding the read by chance in any masked reference sequence
or   any   region   of   the   reference   sequence   coded   as   ‘s,   and   where  is the sum over all the
alignments found plus a factor for chance alignments calculated using the usable read and genome
lengths.
The term allows for the fact that a fragment could have been sourced from portions of the
genome that are not represented in the reference sequence. For instance in Human genome build 36
there is approximately 7% of sequence represented by large blocks of ‘s.
A quality score is calculated as , where is the probability of the
alignment given the read and the genome.

Mus musculus (laboratory mouse) Chromosome

GenBank id	chr	length
NC_000067	chr1	197195432
NC_000068	chr2	181748087
NC_000069	chr3	159599783
NC_000070	chr4	155630120
NC_000071	chr5	152537259
NC_000072	chr6	149517037
NC_000073	chr7	152524553
NC_000074	chr8	131738871
NC_000075	chr9	124076172
NC_000076	chr10	129993255
NC_000077	chr11	121843856
NC_000078	chr12	121257530
NC_000079	chr13	120284312
NC_000080	chr14	125194864
NC_000081	chr15	103494974
NC_000082	chr16	98319150
NC_000083	chr17	95272651
NC_000084	chr18	90772031
NC_000085	chr19	61342430
NC_000086	chrX	166650296
NC_000087	chrY	15902555