arXiv subjects

Cesare Furlanello

Publications and source records attributed to Cesare Furlanello.

27 records · Page 2Linked to original sources

arXiv2012

Stability Indicators in Network Reconstruction

The number of algorithms available to reconstruct a biological network from a dataset of high-throughput measurements is nowadays overwhelming, but evaluating their performance when the gold standard is unknown is a difficult task. Here we propose to use a few reconstruction stability tools as a quantitative solution to this problem. We introduce four indicators to quantitatively assess the stability of a reconstructed network in terms of variability with respect to data subsampling. In particular, we give a measure of the mutual distances among the set of networks generated by a collection of data subsets (and from the network generated on the whole dataset) and we rank nodes and edges according to their decreasing variability within the same set of networks. As a key ingredient, we employ a global/local network distance combined with a bootstrap procedure. We demonstrate the use of the indicators in a controlled situation on a toy dataset, and we show their application on a miRNA microarray dataset with paired tumoral and non-tumoral tissues extracted from a cohort of 241 hepatocellular carcinoma patients.

q-bio.MN↗

arXiv2012

mlpy: Machine Learning Python

mlpy is a Python Open Source Machine Learning library built on top of NumPy/SciPy and the GNU Scientific Libraries. mlpy provides a wide range of state-of-the-art machine learning methods for supervised and unsupervised problems and it is aimed at finding a reasonable compromise among modularity, maintainability, reproducibility, usability and efficiency. mlpy is multiplatform, it works with Python 2 and 3 and it is distributed under GPL3 at the website http://mlpy.fbk.eu.

cs.MS↗

arXiv2011

Single-base mismatch profiles for NGS samples

Within the preprocessing pipeline of a Next Generation Sequencing sample, its set of Single-Base Mismatches is one of the first outcomes, together with the number of correctly aligned reads. The union of these two sets provides a 4x4 matrix (called Single Base Indicator, SBI in what follows) representing a blueprint of the sample and its preprocessing ingredients such as the sequencer, the alignment software, the pipeline parameters. In this note we show that, under the same technological conditions, there is a strong relation between the SBI and the biological nature of the sample. To reach this goal we need to introduce a similarity measure between SBIs: we also show how two measures commonly used in machine learning can be of help in this context.

q-bio.QM↗

arXiv2011

Biological network comparison via Ipsen-Mikhailov distance

Highlighting similarities and differences between networks is an informative task in investigating many biological processes. Typical examples are detecting differences between an inferred network and the corresponding gold standard, or evaluating changes in a dynamic network along time. Although fruitful insights can be drawn by qualitative or feature-based methods, a distance must be used whenever a quantitative assessment is required. Here we introduce the Ipsen-Mikhailov metric for biological network comparison, based on the difference of the distributions of the Laplacian eigenvalues of the compared graphs. Being a spectral measure, its focus is on the general structure of the net so it can overcome the issues affecting local metrics such as the edit distances. Relation with the classical Matthews Correlation Coefficient (MCC) is discussed, showing the finer discriminant resolution achieved by the Ipsen-Mikhailov metric. We conclude with three examples of application in functional genomic tasks, including stability of network reconstruction as robustness to data subsampling, variability in dynamical networks and differences in networks associated to a classification task.

q-bio.MN↗

arXiv2011

A machine learning pipeline for discriminant pathways identification

Motivation: Identifying the molecular pathways more prone to disruption during a pathological process is a key task in network medicine and, more in general, in systems biology. Results: In this work we propose a pipeline that couples a machine learning solution for molecular profiling with a recent network comparison method. The pipeline can identify changes occurring between specific sub-modules of networks built in a case-control biomarker study, discriminating key groups of genes whose interactions are modified by an underlying condition. The proposal is independent from the classification algorithm used. Three applications on genomewide data are presented regarding children susceptibility to air pollution and two neurodegenerative diseases: Parkinson's and Alzheimer's. Availability: Details about the software used for the experiments discussed in this paper are provided in the Appendix.

q-bio.MN↗

arXiv2010

An introduction to spectral distances in networks (extended version)

Many functions have been recently defined to assess the similarity among networks as tools for quantitative comparison. They stem from very different frameworks - and they are tuned for dealing with different situations. Here we show an overview of the spectral distances, highlighting their behavior in some basic cases of static and dynamic synthetic and real networks.

q-bio.MN↗

arXiv2010

A unifying view for performance measures in multi-class prediction

In the last few years, many different performance measures have been introduced to overcome the weakness of the most natural metric, the Accuracy. Among them, Matthews Correlation Coefficient has recently gained popularity among researchers not only in machine learning but also in several application fields such as bioinformatics. Nonetheless, further novel functions are being proposed in literature. We show that Confusion Entropy, a recently introduced classifier performance measure for multi-class problems, has a strong (monotone) relation with the multi-class generalization of a classical metric, the Matthews Correlation Coefficient. Computational evidence in support of the claim is provided, together with an outline of the theoretical explanation.

stat.ML↗

arXiv2010

Algebraic Comparison of Partial Lists in Bioinformatics

The outcome of a functional genomics pipeline is usually a partial list of genomic features, ranked by their relevance in modelling biological phenotype in terms of a classification or regression model. Due to resampling protocols or just within a meta-analysis comparison, instead of one list it is often the case that sets of alternative feature lists (possibly of different lengths) are obtained. Here we introduce a method, based on the algebraic theory of symmetric groups, for studying the variability between lists ("list stability") in the case of lists of unequal length. We provide algorithms evaluating stability for lists embedded in the full feature set or just limited to the features occurring in the partial lists. The method is demonstrated first on synthetic data in a gene filtering task and then for finding gene profiles on a recent prostate cancer dataset.

stat.ML↗

arXiv2002

The Dynamics of AdaBoost Weights Tells You What's Hard to Classify

The dynamical evolution of weights in the Adaboost algorithm contains useful information about the role that the associated data points play in the built of the Adaboost model. In particular, the dynamics induces a bipartition of the data set into two (easy/hard) classes. Easy points are ininfluential in the making of the model, while the varying relevance of hard points can be gauged in terms of an entropy value associated to their evolution. Smooth approximations of entropy highlight regions where classification is most uncertain. Promising results are obtained when methods proposed are applied in the Optimal Sampling framework.

cs.LG↗