Butler Scientifics

Anatomic Pathology. Uveal Melanoma

v. 04/09/2020

AutoDiscovery was of great help in choosing the best markers to focus on in our confirmatory phase

What's This Work About?

What's this work about?

Our objective

The fundamental objectives of the group are the study of molecular tumour pathology related to the identification of new diagnostic, prognostic and therapeutic targets.

One of the goals of this study in particular was to assess how the immunohistochemical expression of the markers that are part of the signaling act as independent predictors of the risk of metastasis or variables of global survival.

What Problem Did We Face?

What problem did we face?

We combined up to 48 factors related to demographics, prognosis, biomarkers and clinical follow-up from 101 patients and provided that information to an external biostatistician.

Given the exhaustive exploratory approach of this study, we were forced to invest a significant amount of time and money in technical meetings with the biostatistician in order to discuss each step forward in the discovery process.

Which Was The Contribution of AutoDiscovery ?

Which Was The Contribution Of AutoDiscovery ?

One of the most interesting relationship identified with AutoDiscovery was that patients at stage T3cN0M0 and high levels of expression of p4E-BP1 show a higher index of Ki67 than those with low/mid levels of expression. This relationship supports the idea that mTOR signalling way has a relevant role in the tumoral growth.

In this work we combined classical statistical tools (hypothesis testing) with the automated exploratory analysis of AutoDiscovery

Which Was The AutoDiscovery Contribution?
¿La expresión de conexinas discrimina pacientes con diferente OS?
Which Was The Contribution Of AutoDiscovery ?

Key AutoDiscovery Features

The key features of AutoDiscovery applied to this work were

  • 1 The depth of analysis, which allowed us to identify relevant markers that were potentially associated to particular subsets of our patients (patients at different stages, tumour samples with different levels of regulation of protein synthesis, etc.).
  • 2 The Hypo Booster tool to easily focus on the most relevant factors potentially involved to be tested in a further confirmatory study, and thus minimizing the number of meetings with the biostatistician.
  • 3 The traceability of the results,which enabled an effective integration with classical statistical tools (such as SPSS).
Butler Scientifics

Hi! We hope you liked this study and how AutoDiscovery, our Automated EDA software, helped the researchers of this project unveil hidden relationships in their data.

Please, feel free to contact with us using the link below.

  • Ray Gutiérrez

    Ray Gutiérrez

    CEO

  • Raúl López

    Raúl López

    Data Scientist

  • Juan Martín

    Juan Martín

    Data Scientist

Powered by AutoDiscovery
Butler Scientifics

Next steps

  • Fruitful collaborations

    We have many other successful stories you can read in our website

    Learn more
  • Data Exploration in Biomedicine

    Check out this article about Data Exploration In Biomedical Research in our blog

    Read article
  • We can help

    We can help you applying Automated Exploratory Analysis in your project

    Let's connect
Powered by AutoDiscovery