有兩個都可以在新浪愛問資料

Bioinformatics.For.Dummies.2nd.Ed.2007.pdf

An Introduction

to Bioinformatics Algorithms.pdf

另外看到 Virginia 大學的一些課程

The 2012 Computational Genomics Course has been rescheduled to

November 28 - December 4, 2012

用mothur從*.sff進行數據處理的流程

#create flowgram file from sff from 454 machine

#sffinfo(sff=test.sff,flow=T)

#trim and bin sequences in flowgram

#trim.flows(flow=test.flow,

oligos=test.oligos,bdiffs=0,pdiffs=1,processors=2)

#denoise to remove sequencing errors, create fasta and qual

files

#shhh.flows(file=test.flow.files, processors=2)

#bin by barcode, trim fasta files, remove low quality

sequence

#trim.seqs(fasta=test.shhh.fasta, name=test.shhh.names,

oligos=test.oligos,flip=T,minlength=200,maxlength=500,maxambig=0,maxhomop=8,bdiffs=0,pdiffs=1,processors=2)

#remove redundant sequences

#unique.seqs(fasta=test.shhh.trim.fasta,

name=test.shhh.trim.names)

#align sequences to template 16S rRNA gene

#align.seqs(fasta=test.shhh.trim.unique.fasta,

reference=silva.bacteria/silva.bacteria.fasta, processors=2)

#summarize results so far

#summary.seqs(fasta=test.shhh.trim.unique.align,

name=test.shhh.trim.names)

#determine clear-span region

#screen.seqs(fasta=test.shhh.trim.unique.align,

name=test.shhh.trim.names, group=test.shhh.groups, end=6333,

optimize=start, criteria=85, processors=2)

#remove sequences and cut alignment to clear span region

#filter.seqs(fasta=test.shhh.trim.unique.good.align, vertical=T,

trump=., processors=2)

#remove redundant sequences

#unique.seqs(fasta=test.shhh.trim.unique.good.filter.fasta,

name=test.shhh.trim.good.names)

#pre cluster sequences

#pre.cluster(fasta=test.shhh.trim.unique.good.filter.unique.fasta,

name=test.shhh.trim.unique.good.filter.names,

group=test.shhh.good.groups, diffs=2)

#run ClimeraSlayer to identify potential chimeras

#chimera.uchime(fasta=test.shhh.trim.unique.good.filter.unique.precluster.fasta,

name=test.shhh.trim.unique.good.filter.unique.precluster.names,

group=test.shhh.good.groups, processors=2)

#remove chimeras identified by ChimeraSlayer

#remove.seqs(accnos=test.shhh.trim.unique.good.filter.unique.precluster.uchime.accnos,

fasta=test.shhh.trim.unique.good.filter.unique.precluster.fasta,

name=test.shhh.trim.unique.good.filter.unique.precluster.names,

group=test.shhh.good.groups)

CSHL Computational Genomics - November, 2011 -- Metagenomics

I

This workshop will run from the Unix/MacOS command line.

Students on Mac's can login to their workstations, and start the

Terminal application (in

Applications/Utilitles.

Students on PC's need to login to courses.cshl.edu.

Once in a terminal window, run the script

/ecg/seqprg/scripts/init_meta1.sh. Type:

/ecg/seqprg/scripts/init_meta1.sh

If things have worked properly, you should have the directory

meta1_work in your home directory, and it should contain

several files.

cd meta1_work

ls

To do the analysis, we will run the mothur program to

analyze microbial communities. We are using an example from

http://www.mothur.org/wiki/Schloss_SOP.

To run the mothur program, type:

/ecg/seqprg/bin/mothur

All commands in mothur look like functions, and need to

end with (), for example: help() or

quit().

In this workshop, we will be doing microbial community analysis

on a sample set of 16S rRNA sequences from a human stool sample.

The mothur program will allow us to address the following

questions:

What is the taxonomic makeup of the sample.

How diverse is the community (what is the dynamic range of

abundance); species "evenness", species "richness".

How to compare two microbial community samples.

In real analyses, the sequences must be pre-processed to remove

bar codes, primers, and non-rRNA sequences. This pre-processing has

already been done for the sequences labeled "final.*". The

preprocessing steps are listed here

The first step classifies the sequences taxonomically, and bins

them into clades. This has already been done, producing the files

final.names,final.taxonomy,final.group,final.fasta,

etc. These are linked into your meta1_work directory. In

addition, a time-consuming step produces final.dist.

To characterize the taxonomic makeup, we first cluster sequences

into OTUs (taxa, clades), use make.shared to count

sequence abundance within those OTUs.

The lines below show commands that can be copied and pasted into

the command line of mothur. Lines beginning with

# are comments to explain the steps; you only need to copy

lines that do not begin with #.

#Cluster sequences into OTU -- we are using the command cluster.split to do that as it allows us to cluster

# sequences according a taxonomic level like order or family

cluster.split(fasta=final.fasta, taxonomy=final.taxonomy, name=final.names, taxlevel=3, processors=4)

# The make.shared creates a file that represent the number of times

# that an OTU is observed in multiple samples

make.shared(list=final.an.list, group=final.groups, label=0.03)

# Since some samples might have better coverage, sub-sample to get a dataset

# with the same number of sequences per sample

sub.sample(shared=final.an.shared,size=400)

# assign a consensus taxonomy to each OTU

classify.otu(list=final.an.list, name=final.names, taxonomy=final.taxonomy, label=0.03, cutoff=80)

# Calculate the relative abundance of taxa

phylotype(taxonomy=final.taxonomy, name=final.names, label=1)

# The make.shared creates a file that represent the number of times that an OTU is observed in multiple samples

make.shared(list=final.tx.list, group=final.groups, label=1)

# Since some samples might have better coverage, subsample to get a dataset

# with the same number of sequences per sample

sub.sample(shared=final.tx.shared, size=400)

# assign a consensus taxonomy to each OTU

classify.otu(list=final.tx.list, name=final.names, taxonomy=final.taxonomy, label=1)

Once the sequences have been classified, we can build a tree

that represents the population:

#Here we want to build a phylogenetic tree

sub.sample(fasta=final.fasta, name=final.names, group=final.groups, persample=T, size=400)

dist.seqs(fasta=final.subsample.unique.fasta, output=lt, processors=2)

#Build Tree

clearcut(phylip=final.subsample.unique.phylip.dist)

You can visualize the tree with TreeView (in

/ecg/Applications/TreeView).

To quantify population diversity, collector's curves are

produced:

#Generates collector's curves

collect.single(shared=final.an.0.03.subsample.shared, calc=chao-invsimpson, freq=100)

rarefaction.single(freq=100)

summary.single(calc=nseqs-coverage-sobs-invsimpson)

To compare two populations, one can use a heat map or venn

diagram:

#Generate heatmaps and a venn diagram to compare samples

heatmap.bin(scale=log2, numotu=50)

heatmap.sim(calc=jclass-thetayc)

venn(groups=26-23-28-44)

#PCOA Analysis to Compare samples

dist.shared(shared=final.an.0.03.subsample.shared, calc=thetayc-jclass)

pcoa(phylip=final.an.0.03.subsample.thetayc.0.03.lt.dist)

pcoa(phylip=final.an.0.03.subsample.jclass.0.03.lt.dist)

#Non-metric multidimensional scaling

nmds(phylip=final.an.0.03.subsample.thetayc.0.03.lt.dist)

nmds(phylip=final.an.0.03.subsample.jclass.0.03.lt.dist)

These programs create .svg files, which you should be

able to visualize with a web browser.

總結

以上是生活随笔為你收集整理的计算基因组学需要计算机知识吗,生物信息学——计算基因组学的一些参考书的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。