MML 2010, the International Workshop on Machine Learning and Music, continues a series of workshops related to artificial intelligence and machine learning in music. In this short article the Programme Chairs summarize the content of the workshop.

This paper proposes a new view of pattern discovery in music: inductive querying a corpus for maximally general distinctive patterns. A pattern is distinctive if it is over-represented with respect to an anticorpus, and maximally general distinctive

Music has a long and rich tradition within machine learning.Some of the first applications of early computing machinery were statistical models of melody learned through the analysis of examples.Today machine learning is applied to highly complex music textures and used routinely for music classification, generation and improvisation, music analysis, and performance analysis, to name just a few key applications.Due to the steady increase in available data, most music processing applications can now be supported by some machine learning component.Methods for machine learning in music can be described in terms of their application task, the representation and characteristics of the input data, and the machine learning algorithm used.Though detailed musicological studies will always require clean symbolic data, the classical boundary of audio versus symbolic input is now starting to dissipate as many methods now translate audio data to the symbolic domain where learning takes place, then back again where learned results are applied to the audio signal.For this Special Issue of Intelligent Data Analysis we are pleased to present five articles that cover a wide range of application tasks and music data types.Submission to this Special Issue was solicited by the Editors and each submission was given three reviews by experts in the field, with a second round if necessary, followed by a check of

In this paper we describe a new method for discovering recurrent patterns in a corpus of segmented melodies. Elements of patterns in this scheme do not represent individual notes but rather represent melodic segments that are sequences of notes. A new knowledge representation for segmental patterns is designed, and a pattern discovery algorithm based on suffix trees is used to discover segmental patterns in large corpora. The method is applied to a large collection of melodies, including Nova Scotia folk songs, Bach chorale melodies, and sections from the Essen folk song database. Patterns are ranked using a statistical significance method that integrates pattern self-overlap, length, and frequency in a corpus into a single measure. A musical interpretation of some of the statistically significant discovered patterns is presented.

MOTIVATION: Gene finding remains an open problem well after the sequencing of the human genome. The low gene sensitivity of current methods is a problem for divergent protein families, because fairly accurate exon assemblies are required before sensitive fold recognition algorithms can be applied. This paper presents a new genomic threading algorithm which integrates the gene finding and fold recognition steps into a single process. The method is applicable to evolutionarily divergent protein families that have retained some trace of their common ancestry, number and phase of introns, sizes of exons and placement of structural elements on specific exons. Such conserved structural signals may be visible despite dramatic evolution of protein sequence. RESULTS: The method is evaluated on the family of helical cytokines by cross-validation sensitivity analysis. The method has also been applied to all intergenic regions of the human genome, and an expression and cloning approach has been coupled with the predictions of the method. Two genes discovered by this method are discussed. SUPPLEMENTARY INFORMATION: All data used and the results obtained in the cross-validation analysis are available at http://www.soi.city.ac.uk/~conklin/papers/GT/

MOTIVATION: Gene expression is dependent on two main types of signals; one involving transcription factors which initiates gene transcription, and another which regulates the translation of a nascent mRNA. These post-transcriptional events play an important yet incompletely understood role in regulating gene expression and cellular behavior. Many of the identified cis acting elements for translational regulation occur within the 3'untranslated region (3' UTR), and some have been observed to occur with surprising regularity within certain protein function classes. RESULTS: In this study, we present a new association rule mining method for discovering nucleotide sequence patterns that appear in more sequences than expected within protein function classes. The method is applied to a database of human 3' UTR sequences, and some significant associations between nucleotide patterns and protein function classes are discovered. Among previously identified patterns, the AU-Rich Element (ARE) is found here to occur within the 3' UTR of cytokines, providing statistical validation of an association often reported in the literature. The method has also identified some GC-rich patterns, found to occur within the 3' UTR of homeodomain transcription factors and nuclear proteins. The method should be applicable to many types of regulatory element discovery.

MOTIVATION: Many of the most interesting functional and evolutionary relationships among proteins are so ancient that they cannot be reliably detected through sequence analysis and are apparent only through a comparison of the tertiary structures. The conserved features can often be described as structural motifs consisting of a few single residues or Secondary Structure (SS) elements. Confidence in such motifs is greatly boosted when they are found in more than a pair of proteins. RESULTS: We describe an algorithm for the automatic discovery of recurring patterns in protein structures. The patterns consist of individual residues having a defined order along the protein's backbone that come close together in the structure and whose spatial conformations are similar. The residues in a pattern need not be close in the protein's sequence. The work described in this paper builds on an earlier reported algorithm for motif discovery. This paper describes a significant improvement of the algorithm which makes it very efficient. The improved efficiency allows us to use it for doing unsupervised learning of patterns occurring in small subsets in a large set of structures, a non-redundant subset of the Protein Data Bank (PDB) database of all known protein structures.

The availability of large expressed sequence tag (EST) databases has led to a revolution in the way new genes are identified. Mining of these databases using known protein sequences as queries is a powerful technique for discovering orthologous and paralogous genes. The scientist is often confronted, however, by an enormous amount of search output owing to the inherent redundancy of EST data. In addition, high search sensitivity often cannot be achieved using only a single member of a protein superfamily as a query. In this paper a technique for addressing both of these issues is described. Assembled EST databases are queried with every member of a protein superfamily, the results are integrated and false positives are pruned from the set. The result is a set of assemblies enriched in members of the protein superfamily under consideration. The technique is applied to the G protein-coupled receptor (GPCR) superfamily in the construction of a GPCR Resource. A novel full-length human GPCR identified from the GPCR Resource is presented, illustrating the utility of the method.

An approach to knowledge discovery in complex molecular databases is described. The machine learning paradigm used is structured concept formation, in which object's described in terms of components and their interrelationships are clustered and organized in a knowledge base. Symbolic images are used to represent classes of structured objects. A discovered molecular knowledge base is successfully used in the interpretation of a high resolution electron density map.>