Bioinformatics Lab, Ljubljana

Nejc Ilenič, our student lab member, wins Kaggle competition on paintings and artists. Congratulations!

Painter by Numbers | Kaggle Kaggle is your home for data science. Learn new skills, build your career, collaborate with other data scientists, and compete in world-class machine learning challenges.

New and beautiful artwork in our lab by Agnieszka Rovsnik. Truly BIOlab. :)

Primož, a Googe Summer of Code student, blogs on his tool for learning about machine learning by visualizing probability maps.

Polynomial Classification and Visualization of Classification Probabilities | Orange Blog Orange Data Mining Toolbox

It is so easy to use PCA in Orange: a new video in the Getting Started series

Getting Started With Orange: 09 Principal Component Analysis Dimensionality reduction with principal component analysis. License: GNU GPL + CC Music by: http://www.bensound.com/ Website: http://orange.biolab.si/ Create...

Our lab member Jure Žbontar successfully defended his PhD thesis yesterday afternoon. His work on estimating depth from stereo images with convolutional networks has been widely cited, while his algorithm reached the best accuracy score on three separate data sets last year. He is now continuing his post-doc at NYU under the mentorship of Yann LeCun.

Congratulations to Jure and to many future successes!

So much power here. :) One photo of before and the other of after (with the machine assembled).

We've just recently got a fresh delivery of 30 8TB disks, summing up to about a 200TB of usable storage! Lots of data will find their home here. :)

Go science!

A programming language for living cells New language lets researchers design novel biological circuits.

Call to students: Apply for this year's Google Summer of Code! :)

Orange is a part of this year's Google Summer of Code! Check out all the amazing projects, contribute to open source and earn money! That's what we call an incredible summer! :)

For communication among young European biologists. Exchange your opinions, knowledge and ideas!

Young European Biologists - YEB A non-profit organization made by and for Biology students.

We are learning the kernels at 369.9 km/h.

[1601.04366] Learning the kernel matrix via predictive low-rank approximations

In-house developed algorithm reveals regulatory roles of RNA binding proteins.
Paper available below.

Our newest paper published in Bioinformatics

Orthogonal matrix factorization enables integrative analysis of multiple RNA binding proteins We use cookies to enhance your experience on our website. By continuing to use our website, you are agreeing to our use of cookies. You can change your cookie settings at any time. Find out more

Photo: Željko Stevanić, IFP

Marinka and Blaž's research on learning by fusing heterogeneous data has been named one of the top 10 achievements at University of Ljubljana in 2015.
Congratulations to both!

Najodmevnejši raziskovalni dosežki Univerze v Ljubljani v letu 2015 Tradicionalni Teden Univerze v Ljubljani se je začel s predstavitvijo najodmevnejših raziskovalnih dosežkov Univerze v Ljubljani v letu 2015.

This is a great resource for data mining beginners! We love it!

Top 10 data mining algorithms in plain English | rayli.net Today, I’m going to explain in plain English the top 10 most influential data mining algorithms as voted on by 3 separate panels in this 2007 survey paper.

Earlier today our colleague Marinka Zitnik successfully defended her PhD thesis on Learning by Fusing Heterogeneous Data. Her work on matrix tri-factorization for data fusion is extremely innovative and already considered to be a state of the art in the field. We are very proud to have Marinka in our lab and we wish her all the best in her future scientific adventures!

Finding eight new bacterial response genes in Dictyostelium in just one month? It took way longer to publish our newest paper, but it's finally out there.

Gene Prioritization by Compressive Data Fusion and Chaining

Hosting Marie and Eric from Tilburg University to discuss how to teach data mining to non-statisticians. Eric is proudly standing next to the always serious Ivan Cankar (in the middle). Marie is the lovely lady on the right.

Faculty of Computer Science and Informatics is hosting a closing meeting of CARE-MI project dealing with the regeneration of damaged heart cells for Ischemic Heart Disease (IHD) patients. Participants are actively preparing a final report for the project, but you already can read a number of periodic reports and articles here: http://cordis.europa.eu/project/rcn/94265_en.html

Our hands-on Introduction to Data Mining graduate course at Baylor College of Medicine was carried out in Orange3.

Introduction to Data Mining. Seven lectures. 40 graduate students. All in Orange and visual programming. (Blaž Zupan & Janez Demšar, with huge background support of Bioinformatics Lab, Ljubljana)

Ten more days to our data fusion tutorial at EMBC in Milano. Over 60 attendees have registered already.

[T1/2.3] – Brown 3 Data Fusion of Everything | EMBC15 Home Authors Oral Presentation Guidelines Poster Guidelines Submission Deadlines Submission of Sessions Proposals Author Instructions for 4 Page Papers Author Instructions for One Page Papers Animal/Human Subjects Students Student Activities Student Paper Competition Finalists Student Paper Competit…

The previous riddle was too easy: the name of the institute was in the title of the book at the back. Consider it a warm-up. Now, this one is really hard: this week, Janez is visiting an institute in a country neighbouring the previous one. Which country, city and institute is it? The first correct answer wins an Orange T-shirt. (Lab members don't qualify!) :)

Janez visited an amazing institute close to Paris last week. Our data-mining program Orange could soon get a set of widgets useful in an entirely different area of science. Question: What's the name of the institute? :)

A) Soleil
B) Sorbonne
C) Sothys

Group poster on Orthogonal nonnegative factorization-based analysis of nineteen protein-RNA interaction CLIP data sets presented by Martin at ISMB in Dublin.
http://bubble.fri.uni-lj.si/zp/misc/MartinStrazar_ismb_2015.pdf

Marinka's ISMB poster on Data set selection and sensitivity estimation in collective factor models:

http://helikoid.si/ismb15/zitnik-zupan-ismb15.png

Last week at the ISMB ECCB conference in Dublin Marinka presented her poster on Data set selection and sensitivity estimation in collective factor models, while Tomaž and Martin explained Orthogonal nonnegative factorization-based analysis of nineteen protein-RNA interaction CLIP data sets. Good job guys!

https://www.iscb.org/ismbeccb2015