research

Capturing Screen Image in Agilent Instruments (E4448A Spectrum Analyzer)

by Forrest Sheng Bao http://fsbao.net

You can use other VISA connectivity software on Mac OS and Linux. Here I only explain how to make it in Agilents software running on Windows.

Lately, I am facing a problem on capturing the measurement screen from an Agilent E4448A PSA Series Spectrum Analyzer. So I did some Googling and figured it out. Though I only tested on E4448A, I think it works for all supported Agilent Instruments.

domains, constraints and labeling in Sicstus Prolog solver, finite domain

by Forrest Sheng Bao http://fsbao.net

It's really tricky to program in Sicstus Prolog solver.

Bellow is an example

:- use_module(library(clpfd)).
a(X,Y):-domain([X], 1,4), domain([Y], 4, 6), X+2#>Y, labeling([],[X,Y]).

Results on Sicstus:

| ?- [d].
% consulting /home/grad2/fsbao/d.pl...
% consulted /home/grad2/fsbao/d.pl in module user, 10 msec 72 bytes
yes
| ?- a(X,Y).
X = 3,
Y = 4 ? ;
X = 4,
Y = 4 ? ;
X = 4,
Y = 5 ? ;
no

Ref: Constraint Logic Programming over Finite Domains, Sicstus Manual, http://www.sics.se/sicstus/docs/3.7.1/html/sicstus_33.html

EDFbrowser working with discontinuous EDF+ file

by Forrest Sheng Bao http://fsbao.net

EDFbrowser is an open source EDF Browser. EDF/EDF+ is an open file format for biomedical time series, especially EEG and ECG/EKG signal.

But, the pain is, new (version 1.08 or later) EDFbrowswer "disabled the possibility to open discontinuous files."

I consulted the author, Teunis van Beelen, and got a solution from him.

Genomic localization for small RNAs in our PLoS One paper

by Forrest Sheng Bao http://fsbao.net

If you are looking for a program to do genome localization of small sequences, such as small RNAs, you can use our program as described in this paper: Small RNA Deep Sequencing Reveals Role for Arabidopsis thaliana RNA-Dependent RNA Polymerases in Viral siRNA Biogenesis. PLoS ONE 4(3): e4971. http://www.plosone.org/article/info:doi/10.1371/journal.pone.0004971 Please cite our paper if you use our program in your papers. Here is a short How-to.

My program is written in Python. So you have to install a Python interpreter first. On most Linux distributions, Python is already installed. On Mac and Windows, please go to http://www.python.org/download/ to download the installation program. Python 2.x is recommended coz I didn't test on Python 3.0.

An article about my work on MIT Technology Review Blog

by Forrest Sheng Bao http://fsbao.net

There was an article related to my work on an MIT Technology Review Blog Wednesday. http://www.technologyreview.com/blog/arxiv/23465/

I didn't know this until someone else in my university told me. I guess the author saw a paper I pre-printed at arXiv and felt it was interesting. I appreciate the author to report my work, but I have some words to say regarding the report.

Maybe because I didn't express my ideas very clearly in my paper, some words in the report are not accurate, such as "dramatic impact" or "Nobody." Firstly, there are many similar or close research done by other groups. Secondly, our work is just a very primitive trial, using limited testing methods on very small amount of patient data, only 6 patients.

Connecting to a Linux workstation made easy

by Forrest Sheng Bao http://fsbao.net

Since 2008, I have been helping Dr. Xie, Dept. of Biological Sciences, Texas Tech University on bioinformatics computing. I worked happy with his Post-Doctoral researcher Dr. Qi, who is now with Indiana University. I plan to write down some instructions so that new PhD students can work happy with me as Dr. Qi did.

Workstation software configurations:

enumerate() array elements rather than indexing them in Python

by Forrest Sheng Bao http://fsbao.net

Recently, I am debugging two Python programs about graphs. I get very confused in many times, whether I am visiting an array element or the index of the array element. For example, I need to get the vertex number by indexing an array, and use that vertex number to index another array.

So, my friend Lay Yuan in Arizona State University told me a trick, using enumerate() in Python. This makes things much much more easier. I guess you are smart enough so this example can explain every thing to you

>>> for i,arr in enumerate([6,3,2,3,4]):
...     print i, arr
...
0 6
1 3
2 2
3 3
4 4

bucket cross-validation? subject-wise k-fold cross-validation?

>>>>>>>>>>>>>>>>>>>>

Update: If you are looking for the Petition to ABC for Linux plug-in of online video streaming service "full episodes," please follow this link http://narnia.cs.ttu.edu/drupal/node/138 I updated my site and changed story numbers.

===================================

by Forrest Sheng Bao http://fsbao.net

This is an article about machine learning and biomedical engineering. And you will need MATLAB and MATLAb Neural Network Toolbox to run my demo code.
An easier official example on using MATLAB NN Toolbox is here: http://www.mathworks.com/access/helpdesk/help/toolbox/nnet/radial_4.html#8370

Dumping Cross-Validation Result out from libsvm

by Forrest Sheng Bao http://fsbao.net

Libsvm (http://www.csie.ntu.edu.tw/~cjlin/libsvm/) is a great SVM implementation, made by a group in National Taiwan University, Taiwan.

But you can't see the cross-validation result, by default. Now I am gonna do a multi-source boosting and I am gonna compute the sensitivity and specificity. So I need the result of each weak classifier.

What to do? Well, libsvm is open sourced! You just need to simply modify the svm-train.c file. Following discussion is based on libsvm v. 2.88, released on Oct. 30, 2008

On the power of molecular computing

by Forrest Sheng Bao http://fsbao.net

Eric You Hsu, Stephen Gang Wu and I spent few hours tonight to go thru these papers about DNA computing. You can see the property of DNA molecules to do some fancy computing stuff. Looks like DNA computing has some potential on logical constraint problems, many of which are NP-complete programs. Thus, they are tough to computer scientists.

Looks like they are all focusing on SAT problems. We are look forward to some better ideas.

Few weeks ago at MIT, when I was at iGem 2008, I saw a fancy research conducted by a joint group in Missouri. They implemented the hash algorithm by controlling the expression of fluorescent genes. I talked to them and wanna do more investigation on this topic.

Syndicate content