BOL: Related items

Anders Krogh Lab

Mon, 30 Sep 2013 19:07:40 -0500

In a lot of my work in bioinformatics, I have been using hidden Markov models (HMMs). As a postdoc with David Haussler at UCSC we developed the so-called profile HMMs (refs). Since then I have applied HMMs to membrane proteins (refs) and gene identification (refs) and have worked on methods for such things as discriminative estimation of HMMs (refs) and alternative decoding algorithms etc. (refs).

Now my main interests are in gene regulation, where we work on promoter analysis; non-coding RNA, where miRNAs and structure prediction are the main areas; and protein structure, where the group is working on methods for structure prediction from sequence. To read more about these topics, please see the research pages.

Lab page @ http://wiki.binf.ku.dk/User:Krogh

Molecular Bioinformatics Lab (MBL)

Tue, 19 Nov 2013 18:23:27 -0600

The main subject of interest in our laboratory is the study of the relationship among sequence, structure, and function in proteins and nucleic acids. Our research can be divided in two major topics:

the study of the sequence-structure relationship
(application -> structure prediction)
the study of the structure-function relationship
(application -> function prediction)

Therefore, anything related to the configuration (sequence) and conformation (structure) in atomic systems of proteins and nucleic acids, and the interaction of these with other elements (function) is of our major interest.

Lab page @ http://melolab.org/mbl/

“On” and “Off” the neuron !!!

Rahul Nayak — Fri, 25 Apr 2014 19:31:13 -0500

Optogenetics is a recent innovation in neuroscience that gives researchers the ability to control the activity of neurons with light. With this powerful tool, researchers are teasing apart the biological basis of memory, behavior, and disease (see “Scientists Make Mice ‘Remember’ Things That Didn’t Happen” and “An On-Off Switch for Anxiety,”). But for the first several years of this technology’s existence, the proteins that scientists added to neurons to make them react to light were only good at activating neurons. That limited researchers’ ability to understand neuronal circuits, sets of interconnected neurons that are thought to control behavior and, when misfiring, to underlie many brain conditions. Problems can arise from any imbalance in circuit activity, whether too much or too little.

Now, two research groups have engineered new optogenetic proteins that can be used to efficiently silence neurons. One of the two new proteins comes from the lab of Karl Deisseroth, a psychiatrist and neuroscientist at Stanford University who helped develop optogenetics as a research tool. His group’s new “off” switch for neurons was created by changing 10 of the 333 amino acids in an existing optogenetic protein, which itself had been engineered by combining natural proteins from green algae. That advance “creates a powerful tool that allows neuroscientists to apply a brake in any specific circuit with millisecond precision,” said Thomas Insel, director of the National Institute of Mental Health, in a released statement. The other new silencing protein, developed by scientists at the Humboldt University of Berlin and collaborators, was created by changing amino acids in the same existing optogenetic protein.

Some researchers are also looking to optogenetics as a potential treatment for patients with a variety of conditions (see “For Mice, and Maybe Men, Pain Is Gone in a Flash,” and “Flipping on the Lights to Halt Seizures”) but there are huge challenges to overcome. The method requires genetic modification of cells to make them light-sensitive. It also requires implanted light sources for all but the shallowest of nerve endings.

MEGADOCK 4.0

Suleman Khan — Thu, 07 Aug 2014 18:08:54 -0500

An ultra–high-performance protein–protein docking software for heterogeneous supercomputers

Summary: The application of protein–protein docking in large-scale interactome analysis is a major challenge in structural bioinformatics and requires huge computing resources. In this work, we present MEGADOCK 4.0, an FFT-based docking software that makes extensive use of recent heterogeneous supercomputers and shows powerful, scalable performance of over 97% strong scaling.

Availability and Implementation: MEGADOCK 4.0 is written in C++ with OpenMPI and NVIDIA CUDA 5.0 (or later) and is freely available to all academic and non-profit users at: http://www.bi.cs.titech.ac.jp/megadock.

Contact: akiyama@cs.titech.ac.jp

Address of the bookmark: http://bioinformatics.oxfordjournals.org/content/early/2014/08/06/bioinformatics.btu532.short

dcGOR

Martin Jones — Sat, 08 Nov 2014 14:54:28 -0600

An R package for analysing ontologies and protein domain annotations has been published in PLoS Computational Biology (http://dx.doi.org/10.1371/journal.pcbi.1003929). The package is distributed as part of CRAN (http://cran.r-project.org/package=dcGOR), and also at GitHub for version control.

The dedicated website is available in http://supfam.org/dcGOR, from which several demos are also provided:

1. Analysing SCOP domains: http://supfam.org/dcGOR/demo-Fang.html

2. Analysing Pfam domains: http://supfam.org/dcGOR/demo-Basu.html

3. Analysing InterPro domains: http://supfam.org/dcGOR/demo-Customisation.html

MutaBind

Jit — Mon, 06 Jun 2016 13:34:09 -0500

MutaBind is a new computational method and server created through NCBI research efforts that maps mutations on a protein structural complex, calculates changes in binding affinity, identifies deleterious mutations and produces a downloadable mutant structural model. http://www.ncbi.nlm.nih.gov/projects/mutabind/index.fcgi/

MutaBind guides you through this process, step by step, starting with selecting a protein complex and inputting PDB code or uploading PDB files. You can also retrieve results with a job ID number, view help documents, and review the MutaBind method and references.

More at http://www.ncbi.nlm.nih.gov/projects/mutabind/index.fcgi/

Protein Subcellular Localization Prediction

Poonam Mahapatra — Thu, 01 Mar 2018 06:20:47 -0600

Assigning subcellular localization to a protein is an important step towards elucidating its interaction partners, function, and potential role(s) in the cellular machinery. Computational tools offer an attractive complement to time-consuming and laborious experimental methods.

http://abi.inf.uni-tuebingen.de/Services/YLoc/webloc.cgi

Address of the bookmark: https://abi.inf.uni-tuebingen.de/Research/Systems%20Biology/protein-subcellular-localization

Tools for Protein-Protein Docking !

Poonam Mahapatra — Wed, 25 Apr 2018 05:15:53 -0500

Predicting the structure of protein–protein complexes using docking approaches is a difficult problem whose major challenges include identifying correct solutions, and properly dealing with molecular flexibility and conformational changes. Following are the tools to predict the structure of protein–protein complexes:

3D-Dock Suite

Global rigid search: FFTShape complementarity and electrostatics

Re-scoring and clustering. Refinement of interface side-chains

3D-Garden

Global rigid search in ensamble

Shape complementarity and Lennard–Jones potential

Side chain and backbone dihedral refinement

DOT

Global rigid search: FFTShape complementarity, electrostatics and VDWNone

Escher NG

Global rigid searchShape complementarity, hydrogen bonds and electrostatic

Integrated in VEGA

GRAMM

Global rigid search: FFT. smooth protein surface representation for soft docking

Shape complementarity and Lennard-Jones potential

Clustering of conformations

GRAMM-X

Global rigid search: FFT. smooth protein surface representation for soft docking

Shape complementarity and Lennard-Jones potentialminimization and re-scoring with multiple filters

HEX

Global rigid search: Fourier correlation of spherical harmonics

Shape complementarity

HADDOCK

Global rigid searchElectrostatic ,VDW and desolvation energy termsMD simulated annealing refinement . Filtering based on external data.

ICM

Global rigid search: Monte CarloEmpirical scoring function

Clustering and selection of conformations. Refinement of interface side-chains and re-scoring

MolFit

Global rigid search: FFTShape complementarity

Clustering of good solutions, filtering using a priori information and small, local rigid rotations around selected conformations

PatchDock

Global rigid searchShape complementarity and atomic desolvation energy

Clustering of conformations

PyDock

Global rigid search:FFTShape complementarity

rescoring by binding electrostatics and desolvation energy

RosettaDock

Local rigid search: Monte Carlo with low and high resolution structure representation levels

Different scoring parameters for the different resolutions

ZDOCK

Global rigid search: FFTShape complementarity, desolvation energy, and electrostatics.

Energy minimization and re-scoringFree for academics

Point to note:

The proper treatment of flexibility in protein–protein docking is still an active field of research. You first should analyzed your proteins in order to define their conformational space and then choose the most suitable method for your docking problem.

Interview Puzzles for Bioinformatician !

Rahul Nayak — Tue, 17 Jul 2018 05:26:18 -0500

These are some of the most famous Interview Puzzles being asked in top tech companies.

Here is a list of Top 25 puzzles which have been asked in top Tech Interview.

Specially for Microsoft Interview Puzzles, you may refer,
Top 15 Microsoft Interview Puzzles
Microsoft Interview Puzzles

Other MOST COMMON Interview Puzzles-
Top 25 Tech Interview Logical Puzzles

Each of the puzzles got repeated a number of times in interviews even for top tech companies

SAMHAR-COVID19 Hackathon

BioStar — Fri, 17 Apr 2020 06:47:10 -0500

Centre for Development of Advanced Computing (C-DAC) under the aegis of the National Supercomputing Mission (NSM), a Ministry of Electronics & Information Technology (MeitY) and Department of Science & Technology (DST) initiative, in association with NVIDIA & OpenACC, announces the SAMHAR-COVID19 Hackathon.

Pandemic outbreak such as Coronavirus outbreak can create huge challenges for the Government and Public Health Officials to gather information quickly and coordinate a response. In such a situation, Artificial Intelligence (AI) can play a huge role in predicting, minimizing and stalling its spread of the virus.

C-DAC has embarked on a program SAMHAR-COVID19 (Supercomputing using AI, ML, Healthcare Analytics based Research for combating COVID19). This opportunity will provide researchers to find solutions for Identifying, Tracking and Forecasting outbreaks of COVID19 and Facilitating Drug Discovery as well.

Participants can update submissions multiple times till the Registration End date.
Each entry can be submitted by a Team comprising of minimum 3 and maximum 5 members (Including the Team Lead).
Participants will have to share the complete work activities with C-DAC. And C-DAC will have right to use the submitted application/solution for SAMHAR-COVID19 programs.
The Award will be given to the Selected/Winning Entry irrespective of the number of members in the Team (members may choose to distribute the amount among themselves).
The decision of the Eminent Jury on the I3 Award will be final and binding.
Award can be for the Team/Company/Institution, as submitted in the Application and cannot be changed later.
Submissions will be considered void if they are in whole or part ill-eligible, incomplete, damaged, altered, counterfeit, obtained through fraud or late submission.

More at https://samhar-covid19hackathon.cdac.in/