BOL: Related items

Five points for bioinformatics software/tools

Jitendra Narayan — Mon, 05 Aug 2013 04:12:32 -0500

In the bioinformatics sector we mostly spend time on computational analysis of huge amounts of data and try to make sense of it, biologically. But, most of the newbie bioinformaticians are faced with dilemma when they receive biological sequence data for the first time. They mostly found confusing over open source, user friendly GUI, and commercial bioinformatics software. Don’t be surprise this is true and also not an easy task to decide, because analytical step is the most crucial part and believe to be the biggest bottleneck in publishing paper in high impact journals. Through this blog I would like to address the pros and cons of both kind of software/tools and try to assist (Hmmm not really, It looks convince) you to make decision on your software selections.

The most common newbie questions are:

Should I try to use these free open source programs? Why are we not trying GUI software for computational analysis? Should I use commercial bioinformatics programs/software?”

1. Let’s be open

We generally think free and cheap are useless. But this concept is not applicable when we discuss open source software. Mostly, the bioinformatics software is developed by highly competitive biological programmers who believe in open sharing of knowledge. They come under Open Bioinformatics Foundation or O|B|F which is a non-profit, volunteer run organization focused on supporting open source programming in bioinformatics. The best part about open source tools/software is that they’re free to download the source code and read exactly what the program does. If you are so inclined, you can view all of the parts of the program and see the logical flow of the pipeline. In addition, open source makes an excellent learning tool for any beginning bioinformatician. Moreover, you can modify existing open source programs to deal with cutting-edge problems or to customize your pipeline. Apart from your computational and analysis work, most of the reviewer also prefers the open source based results so that they can validate the results if validation required.

2. Code headache

As a bioinformatician you are supposed to know the basics of programming languages, and if you are not good at it, then please learn it as soon as possible because you are not a bio-analyst but biological programmers. The open source programs usually lack dedicated service and support teams (often because they were the product of an overworked doc/postdoc!) so you are responsible for troubleshooting your own errors most of the time. We commonly receive the HELP email to support and assist to setup the pipeline; you can also find this kind of request on any QA forum. I personally believe this coding horror brings the biggest downside of open-source programs; where you need some programming skills in order to implement the program in your pipeline. But, if you are not able to fix the pipeline and modify the open source code according to your requirements them you should re-think on your bioinformatician name tag!!!

3. Dive into the codes

Some of the biologist turn bioinformatician says “if you can do the same thing with commercial software then why to get migraine with weird codes”, well this statement looks to me that guys are keen to learn swimming but still don’t like to get wet. If you are still using paid software and doing your work by customer support and clicking some of the well-designed GUI button then perhaps you are not interested in learning and trying new and challenging bioinformatics works. You are missing the basic flavour of bioinformatics. Let’s dive into the coding world, I am sure your will enjoy it. I recommend your to swim freely in code’s sea, and enjoy the journey; do not merely watch it from the outside.

4. Paid does not mean better

The bioinformatics company which are specializes in bioinformatics solutions develop well designed/packed, user friendly software by using a large number of specialised scientist, programmers and support staff. They also provide good services to accomplice your biological analysis work. This means that if you hit a ‘snag’ with your data, help is likely only a phone call away! These companies price their products competitively against the cost of a dedicated bioinformatician. You may be able to afford the program, but not the additional staff! Additionally, most of the functionality that you need in your analysis is already coded into the program. Need to plot a graph? Just click this button right here. It is that easy. But, as a bioinformatician this is not generally well encouraged approach in biological analysis work, because the software is not available to everyone and your data can’t be validated. Moreover, there is very less chances that anyone will repeat your work or love to do similar kind of research (because not all the labs in the world are rich like yours).

5. Take a caution

In biological analysis work, in which you deal GB/TB of data are having maximum chances of getting errors, so please be careful and always cross check your data before coming to any conclusion. Even an error in two line code can alter your entire analysis and display weird results. Some of the scientist blindly believes on commercial software, which is entirely wrong. Using proprietary tools does not absolve you of the need to actually read and research the type of analysis that you are doing. This is particularly true in the case of genome assembly and annotation.

At the end, I would like to tell only one think that open source solutions allows you to do more cutting edge analysis than the commercial tools. So let’s go for it.

Disclaimer:

This is my personal view. I have nothing to do with any company or open source community. The views expressed on these pages are mine alone and not those of my current/past employers. I do reserve the right to remove comments left by spammers or off-topic comments.

Linux for bioinformatician !!!

Rahul Nayak — Thu, 13 Mar 2014 16:59:26 -0500

Linux, free operating system for computers, provides several powerful admin tools and utilities which will help you to manage your systems effectively and handle huge amount of genomic/biological data with an ease. The field of bioinformatics relies heavily on Linux-based computers and software. Although most bioinformatics programs can be compiled to run. If you don’t know what these no so user-friendly tools are and how to use them, you could be spending lot of time trying to perform even the basic admin tasks. The focus of this linux series is to help you understand system admin as well as basic tools, which will help you to become an effective bioinformatician and computational biologist.

For knowledge about Linux and their importance amongst bioinformatician plesae read this article "An introduction to Linux for bioinformatics" by Paul Stothard.

Linux cheat sheet at http://bioinformaticsonline.com/file/view/87/linux-cheat-sheet

Please browse for futher useful linux pages on right hand side ...

Spines

Jitendra Narayan — Tue, 09 Feb 2016 05:07:15 -0600

Spines

Spines is a collection of software tools, developed and used by the Vertebrate Genome Biology Group at the Broad Institute. It provides basic data structures for efficient data manipulation (mostly genomic sequences, alignments, variation etc.), as well as specialized tool sets for various analyses. It also features three sequence alignment packages: Satsuma, a highly parallelized program for high-sensitivity, genome-wide synteny; Papaya, an all-purpose alignment tool for less diverged sequences; and SLAP, a context-sensitive local aligner for diverged sequences with large gaps.

Access Spines here.

http://www.broadinstitute.org/science/programs/genome-biology/spines

Address of the bookmark: http://www.broadinstitute.org/science/programs/genome-biology/spines

Awesome bioinformatics pipelines !

Jitendra Prajapati — Wed, 30 Mar 2016 21:50:41 -0500

A curated list of awesome pipeline toolkits ...

https://github.com/pditommaso/awesome-pipeline

Address of the bookmark: https://github.com/pditommaso/awesome-pipeline

Tools and Method for Haplotype phasing !

Manisha Mishra — Fri, 04 Sep 2020 20:41:40 -0500

Huge amounts of genotype data are being produced with recent technological advances, both from increasingly comprehensive and inexpensive genome-wide SNP microarrays and from ever more accessible whole-genome and whole-exome sequencing methods. The vast amount of knowledge contained in these results, however, is best exploited through phased haplotypes, which classify the alleles co-located on the same chromosome. Since sequence and SNP array data normally take the form of unphased genotypes, one does not specifically observe which of the two parental chromosomes, or haplotypes, falls on a specific allele. Fortunately, new advances in both computational and laboratory methods promise improved determination of haplotype phase. Following are useful tools :

Arlequin: http://cmpg.unibe.ch/software/arlequin3/

BEAGLE: http://faculty.washington.edu/browning/beagle/beagle.html

fastPHASE: http://stephenslab.uchicago.edu/software.html

GENEHUNTER: http://linkage.rockefeller.edu/soft/gh/

The Genome Analysis Toolkit:

http://www.broadinstitute.org/gsa/wiki/index.php/The_Genome_Analysis_Toolkit

IMPUTE2: https://mathgen.stats.ox.ac.uk/impute/impute_v2.html

MACH: http://www.sph.umich.edu/csg/abecasis/MACH/

MERLIN: http://www.sph.umich.edu/csg/abecasis/Merlin/

PHASE: http://stephenslab.uchicago.edu/software.html

PL-EM: http://www.people.fas.harvard.edu/~junliu/plem/

“Read-backed phasing” algorithm: http://www.broadinstitute.org/gsa/wiki/index.php/Read-backed_phasing_algorithm

SHAPE-IT: http://www.griv.org/shapeit/

Finding Patterns in Biological Sequences

Jit — Thu, 22 Dec 2016 10:30:49 -0600

In this report we provide an overview of known techniques for discovery of patterns of biological sequences (DNA and proteins). We also provide biological motivation, and methods of biological verification of such patterns. Finally we list publicly available tools and databases for pattern discovery. On-line supplement is available through http://genetics.uwaterloo.ca/∼tvinar/cs798g/motif.

Address of the bookmark: http://engr.case.edu/li_jing/papers/00798gpattern.pdf

Bioinformatics tools to detect horizontal gene transfer (HGT) in genomes

Jit — Fri, 02 Mar 2018 04:56:23 -0600

Horizontal gene transfer (HGT), the “non-sexual movement of genetic material between two organisms” , is relatively common in prokaryotes and single-celled eukaryotes, but a number of factors combine to make it far rarer in multicellular eukaryotes. In order for a eukaryotic species to gain a gene by HGT, foreign DNA must enter the host nucleus, integrate into the genome, and in more complex organisms it must enter the sequestered germline in order to be transmitted to offspring. Once there, it must not experience strong negative selection, despite potential for genetic incompatibility with the host genome and mismatch between the niche of the donor and the host. Over the longer term, foreign DNA may become “domesticated” in the recipient genome and provide novel function.

Following are the popular tool to detect HGT in genomes:

T-REX / 3.22

HGT detection / download & compile

20525630

RANGER-DTL / 2.0

HGT detection / download binary

22689773

PhyloNet / 3.6.1

HGT detection / download binary

18662388

Jane / 4.01

HGT detection / download binary (!license!)

20181081

TREE-PUZZLE / 5.3.rc16

HGT detection / download & compile

11934758

CONSEL / 0.20

HGT detection / download

11751242

DarkHorse / 1.5 rev170

HGT detection / download & install

17274820

HGTector / 0.2.1

HGT detection / git clone

25159222

EGID / 1.0

HGT detection / download

22355228

GeneMarkS / 4.30

HGT detection / download binary (!license!)

9461475

Protein-Protein Interaction Sites Predictions !

Poonam Mahapatra — Wed, 25 Apr 2018 04:53:20 -0500

The study of Protein–Protein Interactions (PPIs) has a crucial role in biology, medicine and the pharmaceutical industry. PPIs can be investigated from two aspects: The interaction partners of a specific protein and the amino acid residues participating in a given PPI. Information about a protein’s interaction partners allows scientists to construct protein interaction networks, such as signaling pathways, which in turn facilitate the understanding of many biological and clinical observations.

Following are the list of tools commonly used to PPIs predictions:

Protein-Protein Interaction Sites

PPISP

A consensus neural network method for predicting protein-protein interaction sites

HOMCOS

A server to predict interacting protein pairs and interacting sites by homology modeling of complex structures

HotPOINT

Prediction of protein interfaces using an empirical model

ISIS

Prediction of interaction hotspots from sequence

KFC server

Automated decision-tree approach to predicting protein-protein interaction hot spots

meta-PPISP

A meta server for predicting protein-protein interaction sites. meta-PPISP is built on three individual web servers: cons-PPISP, PINUP, and Promate

ODA

Identification of optimal surface patches with the lowest docking desolvation energy values

PINUP

Protein binding site prediction with an empirical scoring function

Other Sites (DNA, RNA, Metals)

CHED

Web server for predicting soft metal binding sites in proteins

DBD-Hunter

A knowledge-based method for the prediction of DNA-protein interactions

DISPLAR

Given the structure of a protein known to bind DNA, the method predicts residues that contact DNA using neural network method

iDBPs

Predicts DNA binding proteins for proteins with known 3D structure.

PFplus

A tool for extracting and displaying positive electrostatic patches on protein surfaces which can be indicative of nucleic acid binding interfaces.

MITObim - mitochondrial baiting and iterative mapping

Rahul Nayak — Tue, 08 May 2018 04:15:25 -0500

This document contains instructions on how to use the MITObim pipeline described in Hahn et al. 2013. The full article can be found here. Kindly cite the article if you are using MITObim in your work. The pipeline was originally developed for Illumina data, but thanks to the versatility of the MIRA assembler, MITObim supports in principle also data from the Iontorrent, 454 and PacBio sequencing platforms.

Below you can find a few basic tutorials for how to run MITObim and I encorage you to give them a try with the testdata that comes with this Repo, just to make sure everything is running smoothly on your system. It'll only take a few minutes and will potentially safe you a lot of time down the line.

I provide further examples here as Jupyter notebooks. Get in touch if you feel like sharing your particular MITObim solution and I'd be happy to put it up here, too!

Address of the bookmark: https://github.com/chrishah/MITObim

mmgenome: Tools for extracting individual genomes from metagneomes

Jit — Thu, 09 Aug 2018 17:41:17 -0500

The mmgenome toolbox enables reproducible extraction of individual genomes from metagenomes. It builds on the multi-metagenome concept, but wraps most of the process of extracting genomes in simple R functions. Thereby making the whole process of binning easy and at the same time reproducible through the Rmarkdown format.

The mmgenome R package also facilitates effortless integration with additional data sources and hence should not be seen as "yet another binning method", but rather a package to integrate different binning strategies.

All functions in the mmgenome R package has associated documentation, check it out in R by e.g. ?mmplot.

Address of the bookmark: https://github.com/MadsAlbertsen/mmgenome