The Ludesi 2D PAGE image analysis blog - featuring the REDFIN software

The new REDFIN time licenses and why you should buy them now

2010-12-16T09:41:25Z

In celebration of the festive season, but also of our new REDFIN time licenses, we are giving 10% off all orders of time licenses made before Dec 31!

Time licenses give you full access to REDFIN for a specified period of time, making them ideal for labs that run one or more large 2D gel projects a year. You can either buy a yearly license for €2500, or monthly licenses for €350 each and use them as you need them. Together with the great pay-per-gel system in REDFIN this allows you to keep your costs to a minimum whilst using state-of-the-art analytical software in your research.

If you move fast, you can take advantage of this great end-of-year special and save €250 on every yearly license, or €35 on every monthly REDFIN license you buy!

All servers back up and running again

2010-11-16T18:26:56Z

The issues of the last few days have all been looked into and fixed where needed. We are therefore happy to report that all servers are up and running again and REDFIN is fully functional for all accounts.

Apologies to those of you who have been affected by this outage!

Server Issues

2010-11-08T16:32:03Z

This afternoon problems with some of the servers hosting REDFIN have occurred. If you are affected, you will notice this by your status saying “offline” on your REDFIN start page.

Whilst the servers are down you will not be able to download new results to your REDFIN account, or bring any completed results back into analysis.

However, you are able to analyze your gels (just not download the results), as well as access any results you have already once opened on your computer.

The company responsible is currently looking into the problem and will hopefully have this fixed as soon as possible.

Good-bye Ludesi Analysis Center. Hello REDFIN!

2010-11-02T14:13:56Z

These last few months have been a roller-coaster ride for us here at Ludesi! New developments have occured and rumours have been circulating and it’s time to put the record straight.

Here is a run-down of what is new and what stays the same:

What is new with: Ludesi?

Due to significant shifts on the proteomics and electrophoresis market in recent years and with strained budgets from the global financial downturn, the overall demand for image analysis services has gradually declined. Because of these developments on the market, Ludesi decided in September to close down its well-known image analysis service operations. However, the rights to the award-winning REDFIN software have been acquired by key staff from the old Ludesi team and remains under management of former CEO Ola Forsstrom-Olsson to ensure continuity for staff and customers.

If you have heard rumours that you will no longer be able to access REDFIN, you may safely ignore these. Unfortunately, such rumours have been the result of one of our competitors sales strategy in order to scare you into making hasty purchases from them. Don’t let yourself be tricked in this way.

What is new with: REDFIN?

Time-based licenses are now available for large volume users.
One of the great advantages of the REDFIN 2D gel image analysis software is the convenient pay-per-gel option. For those of you analyzing only a few dozen gels per year, this is by far the cheapest solution for getting your hands on powerful image analysis software with all the features. However, we realize that once you start running larger volumes of gels, the math doesn’t quite work as well. This is why we are now starting to introduce competitive time-based licenses for REDFIN, similar to software licenses. Contact us at info@ludesi.com for more information or simply keep your eyes open for more detailed information that we will be sending your way soon.
The Analysis Service options Basic and Pro have been removed.
In the last two years the analysis part of the REDFIN software has been developed to a very high standard. This has not only made REDFIN into one of the market-leading 2D gel image analysis software, but has also decreased demand for the analysis service options Basic and Pro. In response to this, we have, with a heavy heart, decided to close down the Analysis Center and fully focus on REDFIN as a 2D gel image analysis software. If you haven’t yet tried out the easy analysis wizard in REDFIN, you can find plenty of step-by-step instructions and video tutorials on our website.

Trying to pay by credit card?

We are in the process of switching over to PayPal for credit card payments, which means you are not able to choose the credit card payment option in REDFIN for the moment. If you would still like to pay using your credit card, please contact us at support@ludesi.com stating the number of analysis credits you wish to buy. You will then receive an email with a PayPal link and you will receive your image analysis credits shortly after. We apologize for the inconvenience and hope to have this incorpororated into REDFIN again soon.

REDFIN now also automatically tests your 2D gel data using the non-parametric Mann-Whitney procedure

2010-07-06T09:03:44Z

A few weeks ago you voted on the top features you’d like to see in REDFIN next and your highest priority was to have more types of statistical hypothesis testing available at your fingertips.

So we got busy and not even three weeks later we’re proud to announce that in addition to the ANOVA/Student’s t-test, your 2D gel image analysis data is now also automatically evaluated using the Mann-Whitney test.

But let’s take a step back and look at the reasons behind implementing more procedures for statistical significance testing and what the deal is specifically with Mann-Whitney.

For most of us, the topic of statistical significance testing is a necessary evil we have to face in order to evaluate the data from our research appropriately. The ongoing discussions of “right” and “wrong” tests for research data and the absence of clear guidelines, especially in the area of 2D gel-based proteomics, can leave many of us confused and uncertain about how to statistically evaluate our results in the most correct way.

So why is it so hard to decide on the most appropriate test to use?

The biggest problem with 2D gel experiments (from a statistical point of view) is the low sample number. The average sample size in our area of research is maybe two groups with 4 samples per group, so 8 samples total. And I am talking about biological replicates here not technical replicates… (a whole different hornets nest that we poked at a little bit in this post and this post). Even if you deserve my virtual pat on the back by having run, say, a 60 gel experiment (anyone doing 2D gel based proteomics knows what an impressive feat that is!) it is still not worthy of a statisticians smile, who consider decent sample numbers to start at 100+.

The problem is that statistical significance testing (the thing that gives us our beloved p-value) is built around assumptions of distribution, in most cases it is assumed that a population follows a normal, or bell-shaped, distribution. In the case of a normal distribution, we should always employ so-called parametric tests such as the Student’s t-test or ANOVA, as they offer us the most power and sensitivity. But as my favorite online statistics site StatSoft points out:

For many variables of interest, we simply do not know for sure that this is the case. For example, is income distributed normally in the population? — probably not. The incidence rates of rare diseases are not normally distributed in the population, the number of car accidents is also not normally distributed, and neither are very many other variables in which a researcher might be interested. (read more)

What about 2D gel data?

Unfortunately, 2D gel data falls in the category of ” simply do not know for sure how it is distributed”. The reason for this is that we never have enough samples to use the tests that would confirm if the data is normally distributed or not. So it could be that our data is normally distributed, but we simply don’t know for sure.

In cases where we are not sure what type of distribution we are dealing with, we can fall back on so-called non-parametric tests. These tests don’t assume a normal distribution (although they do assume other things…) and you can normally always find a non-parametric equivalent to any parametric test out there. The down-side of our non-parametric sidekicks is that they are not as powerful as their parametric super-hero equivalents. What makes them more generally applicable unfortunately also reduces their sensitivity. So solely relying on non-parametric test could mean that we miss out on interesting findings in our experiments.

The best solution seems to be to use multiple tests together with your own good judgement as to which spots are worth investigating further. Ultimately, any interesting changes in expression always need to be validated downstream using other experimental methods and the statistical evaluations we do simply help us narrow down our search considerably and give us confidence in our results.

This is why we have implemented the non-parametric Mann-Whitney test in REDFIN in addition to the ANOVA/Student’s t-test that is being performed already.

The REDFIN software displays p-values from the non-parametric Mann-Whitney test in addition to p-values from an ANOVA.

You can see both p-values side-by-side and filter/sort on them either individually or together, depending on your specific experiment and preferences. Note, that the Mann-Whitney test is a ranking test so don’t expect the p-values of an ANOVA and a Mann-Whitney to be the same.

And finally, as with any type of statistical evaluation, never forget that we are dealing with probabilities and not with ultimate answers.

The Mann-Whitney test is the first of several additional hypothesis test we are planning to implement, but remember that you should not feel restricted to the tests we are offering directly in the REDFIN software, as you can easily export all normalized spot volumes for further statistical processing in a statistical software package of your choice.

“Advisable routines for working with your 2D gel results” webinar now open for registration

2010-06-29T17:21:53Z

The dates for the third webinar in our series of free educational webinars are now out and seats are already being reserved. This time we will be discussing:

“Working with your 2D gel image analysis results: advisable routines for controlling, mining, and exporting your data.”

For most 2D gel experiments, data mining will be the final step before moving on to protein identification using mass spectrometry.

The statistical tests and exploratory procedures you use will determine which spots to excise from the gel and applying the correct ones will minimize the risk of obtaining ID’s for proteins that are actually irrelevant to the biological question you where asking.

This webinar will teach you the basics of successful data mining for 2D gel data, as well as more advanced methods for exploring your data further. For example you will learn how to confirm that your gels are indeed clustering into their experimental groups, that there exist no outliers in your data, and how to investigate the influence of other biological or experimental factors within the same analysis project. Basically you will be kitted out with the tools you need to interpret your data correctly and find the expression changes you are really after.

To round it all off, the webinar will also briefly cover the main points of exporting your data from the REDFIN software for use in downstream processing, and show you how you can interactively collaborate on experiments with your colleagues, wherever they may be on planet earth.

Webinar time and dates, as well as a registration link can all be found on the Ludesi webinar page.

REDFIN users voted on their next preferred feature in the REDFIN software

2010-06-17T14:40:43Z

Last week we asked REDFIN users to tell us which feature they think our development team should work on next.

We have tons of ideas for what to do next in REDFIN but we want to allow the people actually using the software to help shape it into the best possible software for their 2D gel analysis needs.

So we sent out a poll suggesting 6 possible features (based on previous customer feedback) that we could start work on straight away.

The choice was

More statistical hypothesis testing (Mann-Whitney, Wilcoxon)
More descriptive statistics
Display of q-values in addition to p-values
Extended annotation capabilities
Customization of PDF reports
MW/pI estimation of spots

And here are the results:

Starting with the implementation of the Mann-Whitney non-parametric test, we will be implementing more methods for statistical hypothesis testing in REDFIN over the coming weeks.

And like with all new features or bug fixes in REDFIN, your local version of the REDFIN software will automatically and at no cost update itself to the newest version, as soon as there is one.

So keep your eyes open for some interesting updates coming your way soon…

Gel IQ featured on Genetic Engineering & Biotechnology News “Best Of The Web”

2010-06-16T10:43:26Z

Gel IQ has been picked up by Genetic Engineering & Biotechnology News (GEN) columnist Taralyn Tan. Taralyn studies nociceptor development and neuropathic pain in a Harvard neurobiology laboratory at the Dana Farber Cancer Institute and as a side-project she reviews websites of interest within the ‘OMICS’ field. The ones she likes get featured on GEN “Best of the web” section and this is were you can also find her review of Gel IQ.

These last few weeks (and especially following our webinars on 2D gel data quality) many of you have been downloading the free Gel IQ tool and scoring the quality of your 2D gel image analysis data. Because Gel IQ pin-points exactly if and where there are any weaknesses in your analysis, it forms an ideal intermediate step between image analysis and data mining/protein identification. Especially with the on-going efforts to standardize and improve proteomics data quality, I think that this smart little tool (did I mention it’s completely free?) should be found in all our 2D gel labs.

So thanks Taralyn for the mention!

Investigating Systems Biology: Dr. Pascale Jackers combines proteomic and transciptomic data to study biological interactions

2010-06-08T09:01:35Z

In today’s customer profile I have the pleasure of introducing you to Dr. Pascale Jackers. Dr. Jackers works as a Senior Scientist in the well-established, multi-disciplinary Mass Spectrometry Laboratory situated at the University of Liège in Belgium.

Above image: Dr. Jackers Pascale and her colleagues at the Mass Spectrometry Laboratory at the University of Liège

Under the direction of Prof. Edwin DePauw, the team uses primarily mass spectrometric techniques to research in the fields of analytical chemistry (i.e. quantification of pollutants or illicit products), medical chemistry (analysis of proteins expressed in a given tissue or a cell culture, identification and quantification of new biomarkers), biological and pharmaceutical chemistry (analysis of the interaction between a new molecule and its target protein or DNA), biophysics and structural chemistry (structure of supramolecular assemblies).

To complement the labs mass spec approaches, Dr. Jackers also uses 2D PAGE to investigate interactions within a biological system with special focus on post translational modifications (PTM’s). Her latest projects with Ludesi involved the analysis of large-scale time-course experiments using DIGE.

Read below for Dr. Jackers’s answers to our seven customer profile questions:

How would you describe the overall goal of your research?

My foremost goal is to understand systems biology using data from proteomics (including PTMs) and transcriptomics when available.

What is your goal specifically for your 2D gel based research?

To obtain accurate data with correctly identified proteins. Such data could be used as a basic support to study PTMs and systems biology.

What type of samples are you primarily working with?

Mouse brain tissue

What have you achieved by using 2D gels?

We are still in the process of completing this project, so you could say our goals have not been completely achieved yet.

What type of Ludesi analysis do you primarily use (REDFIN Solo, Basic, or Pro)?

Ludesi Pro analysis

What made you choose Ludesi’s analysis service?

The main reason is the all-to-all matching algorithm employed in the Ludesi Analysis Center, where reference-free matching is performed. Statistically, that kind of algorithm leads to a better accuracy compared to other software that require the researcher to select a reference gel. Another reason is the Combined Correctness algorithm that is used by an independent team of experts to quality control the analysis of the gels.

How does using the analysis service help you in your work?

I’ve got higher confidence in the generated results.

Read more about Dr. Jackers’s work in some of her recent publications:

VALAPODYN: Validated Predictive Dynamic Model of Complex Intracellular Pathways Related to Cell Death and Survival
A. Depaulis, E. Vafiadaki, P. Jackers, D. A. Arvanitis, O. Kel-Margoulis, E. Wingender, T. Vujasinovic, D. Greenberg, H. Soreq, E. De Pauw, J. B. Dumas Milne Edwards, D. Sanoudou.
European Journal of Human Genetics (2007) 15 (suppl. 1), 307

Developing predictive dynamic models of complex intracellular networks for neurological disease
E. Vafiadaki, A. Depaulis, P. Jackers, D. Arvanitis, O. Kel-Margoulis, E. Wingender, T. Vujasinovic, D. Greenberg, H. Soreq, E. DePauw, J. B. Dumas,M. Edwards and D. Sanoudou.
FEBS Journal (2008) 275 (suppl. 1), 206.

Click here to see all customer profiles

REDFIN 2D gel image analysis solution chosen by Bio-Rad to strengthen their electrophoresis portfolio!

2010-06-08T09:03:45Z

It’s now official: Bio-Rad Laboratories, Inc., one of the largest global manufacturers and suppliers of life science products and the market-leader in 2D gel electrophoresis products has incorporated REDFIN into their electrophoresis portfolio. Wow!

Apart from being a huge recognition for the REDFIN software and the Ludesi 2D gel image analysis service (which is accessible through the REDFIN software), it will also raise awareness of this great product on the global proteomics market and make it even more accessible to researchers world-wide.

And in case you were wondering, REDFIN will continue to be maintained, developed and managed by Ludesi, but will now also benefit from the impressive distribution channels of Bio-Rad.

There are several reasons why Bio-Rad chose the REDFIN software/service solution above all other 2D gel image analysis software on the market. Katy McGirr, Product Manager, Lab Separations Division, summarizes the key features that convinced them of the superiority of REDFIN in their recent press release:

“With REDFIN 2-D analysis service, researchers no longer need to perform the tedious tasks of manually counting and aligning spots on their gels. REDFIN helps researchers who are getting started with 2-D analysis, providing them with guidance through the process as well as giving additional confidence in their final results. The REDFIN 2-D analysis service will save 2-D gel electrophoresis users time by offering them an easy-to-use solution that yields faster results and peace of mind by having the analysis performed by experts. And with the ‘pay-as-you-go’ nature of the service, REDFIN analysis is a low-risk way for researchers to get started in 2-D gel experiments.”

For Bio-Rad the key benefits of REDFIN are:

1. Great design. An easy-to-use and navigate software interface that guides the user through pre-defined workflows and data mining pages.

2. Versatility. The versatility of using any 2-D stain, including DIGE and protein-specific stains.

3.Feature-rich. The convenience of performing advanced statistical calculations on the results within the REDFIN software.

4. Added data security. The results are password protected and backed up by the REDFIN service.

5. Collaboration functionality. An easy way for researchers to share information interactively via email invite.

It goes without saying that this exciting partnership will also result in some great perks for you, the end-user, but I’ll save that one for my next blog post…