The global demand for Artificial Intelligence (AI) talent is growing at a rapid pace, leading to significant AI skills shortages. This experience report describes the creation of a new and uniquely joint-faculty AI concentration within our existing Master's program, characterized by its applied research internship in industry. We describe the experience of creating a multi-disciplinary AI program through broad consultation across two of the largest faculties in our institution. The new concentration has been well-received by the collaborating faculties and with industry partners; its popularity with applicants has resulted in admitting exceptional candidates from around the world and becoming the most popular concentration in the program. We offer this experience report in the hope that it may serve as a model for other practitioners who are considering navigating the creation of a joint-faculty concentration within a graduate program, especially in the popular field of AI.

BACKGROUND: Reconstructing ECG signals from PPG measurements is a critical task for non-invasive cardiac monitoring. While several public ECG-PPG datasets exist, they lack the diversity found in image datasets, and the data collection process often introduces noise, making ECG reconstruction from PPG signals challenging even for advanced machine learning models. RESULTS: We propose a novel ODE-based method for generating synthetic ECG-PPG pairs to enhance training diversity. Building on this, we introduce CLEP-GAN, a subject-independent PPG-to-ECG reconstruction framework that integrates contrastive learning, adversarial learning, and attention gating. CLEP-GAN achieves performance that matches or surpasses current state-of-the-art methods, particularly in reconstructing ECG signals from unseen subjects. Evaluation on real-world datasets (BIDMC and CapnoBase) confirms its effectiveness. Additionally, our analysis shows that demographic factors such as sex and age significantly impact reconstruction accuracy, emphasizing the importance of incorporating demographic diversity during model training and data augmentation. CONCLUSIONS: Our method produces synthetic ECG-PPG pairs with RR interval distributions closely aligned with their real counterparts and shows strong potential to simulate diverse rhythms such as regular sinus rhythm (RSR), sinus arrhythmia (SA), and atrial fibrillation (AFib). Furthermore, CLEP-GAN demonstrates robust performance on both synthetic and real datasets, achieving near-perfect reconstruction in synthetic settings and competitive results on real data. These findings highlight CLEP-GAN's promise for reliable, non-invasive ECG monitoring in clinical applications.

The primary objective of this work was to identify those human phosphorylation sites (phosphosites) that are highly conserved in other species as this may reveal functionally important phosphosites. We wondered whether human phosphosites (e.g. tyrosine, serine or threonine) that are known to be activatory upon phosphorylation, are commonly replaced by a glutamic or aspartic acid residues in other species. This type of alteration might mimic constitutive phosphorylation of cognate proteins in other species, which would indicate that phosphorylation of these sites in humans may confer functionality. To investigate this, we developed an algorithm to identify ortholog proteins in different species for each given human phospho-protein and also predict phosphosites in every extracted ortholog cognate protein. The results demonstrate that relatively few human phosphosites are highly conserved; for instance from about 90,000 human phosphosites, about 75% of these were conserved in mammals, but less than 16% were detected in most model organisms. These extremely well conserved phosphosites did not display any increased preponderance for acidic amino acid substitutions. However, we observed that the most conserved functional phosphosites occurred on threonine phosphosites that were found in protein kinases and these were 8-times more likely to be stimulatory than inhibitory.

In this paper we propose a new algorithm to predict the phosphorylation site specificities of 478 human protein kinases based on the primary structures of the catalytic domains of these enzymes. Existing methods deduce the specificity of a protein kinase through the alignment of the amino acid sequences of phospho-sites targeted by the kinase to generate a consensus sequence or they use machine learning models for recognition. However, for most protein kinases few if any substrates have been experimentally identified by protein sequencing and mass spectrometry. In this work, we used mutual information from a training set of over 200 protein kinases consensus phospho-site sequences and predicted amino acid interactions between kinases and their substrate phospho-sites to generate position-specific scoring matrices (PSSM). The results demonstrate that using our algorithm, knowledge of the primary amino acid sequence of the catalytic domain of these kinases is sufficient to predict their phosphorylation sites specificities and their PSSM matrices.

BACKGROUND: Recent biological discoveries have shown that clustering large datasets is essential for better understanding biology in many areas. Spectral clustering in particular has proven to be a powerful tool amenable for many applications. However, it cannot be directly applied to large datasets due to time and memory limitations. To address this issue, we have modified spectral clustering by adding an information preserving sampling procedure and applying a post-processing stage. We call this entire algorithm SamSPECTRAL. RESULTS: We tested our algorithm on flow cytometry data as an example of large, multidimensional data containing potentially hundreds of thousands of data points (i.e., "events" in flow cytometry, typically corresponding to cells). Compared to two state of the art model-based flow cytometry clustering methods, SamSPECTRAL demonstrates significant advantages in proper identification of populations with non-elliptical shapes, low density populations close to dense ones, minor subpopulations of a major population and rare populations. CONCLUSIONS: This work is the first successful attempt to apply spectral methodology on flow cytometry data. An implementation of our algorithm as an R package is freely available through BioConductor.

We introduce two variants of proper colorings with imposed partial ordering on the set of colors. One variant shows very close connections to some fundamental problems in graph theory, e.g., directed graph homomorphism and list colorings. We study the border between tractability and intractability for both variants.

Homology search is a fundamental tool in genomics and proteomics research and considerable attention has been paid to the development of faster homology finding algorithms. The most efficient homology search algorithms known are based on finding short exact or near exact subsequences and then extending these to form long matches. Traditional tools such as FASTA and BLAST use the consecutive seed model, that is, an exact short match is found and then extended. More recently, PatternHunter, which uses the spaced seeds model, has shown considerably better sensitivity and attention has been focused on the design of optimal seeds for specific data sets. In some cases, the use of multiple seeds has been shown to be necessary. Here we focus on the problem of applying multiple seeds to a dataset in an efficient manner. In particular, when multiple seeds are applied, each seed can result in the same subsequence being found necessitating the removal of duplicates. We propose an efficient data structure that, given a set of seeds and two input sequences, reports the set of matching subsequences (that is, reports matching subsequences without duplications). We show that for any number of seeds, our algorithm is asymptotically optimal when the input sequences are longer than a certain threshold.

Named for the combination of the words DJ and Jamming, the DJammer allows its users to manipulate their music using standard DJ techniques as well as interact with others in virtual jam sessions through the exchange and sharing of multiple music streams. In this paper, we describe the evolution of the DJammer including the interactive process and evaluation which led to the latest prototype. This latest prototype allows its users to "touch" their digital music through single-handed manipulation and control. By allowing both creativity and communication with digital music, the DJammer is a new musical instrument which takes the next step in the evolution of portable music players.

Vehicle routing problems involve the navigation of one or more vehicles through a network of locations. Locations have associated handling times as well as time windows during which they are active. The arcs connecting locations have time costs associated with them. In this paper, we consider two different problems in single vehicle routing. The first is to find least time cost routes between all pairs of nodes in a network for navigating vehicles; we call this the all pairs routing problem. We show that there is an O(log/sup 2/ n) time parallel algorithm using a polynomial number of processors for this problem on a CREW PRAM. We next consider the problem in which a vehicle services all locations in a network. Here, locations can be passed through at any time but only serviced during their time window. The general problem is NP-complete under even fairly stringent restrictions but polynomial algorithms have been developed for some special cases. In particular, when the network is a line, there is no time cost in servicing a location, and all time windows are unbounded at either their lower or upper end, O(n/sup 2/) algorithms have been developed. We show that under the same conditions, we can reduce this problem to the all pairs routing problem and therefore obtain an O(log/sup 2/ n) time parallel algorithm on a CREW PRAM.

The proof of Wagner's conjecture by Robertson and Seymour gives a finite description of any family of graphs which is closed under the minor ordering. This description is a finite set of minimal graphs not in the family; these graphs are called the obstructions of the family. Since the intersection and union of two minor closed graph families is again a minor closed graph family, an interesting question regards computing the obstructions of the new family given the obstructions for the original two families. It is easy to compute the obstructions of the intersection, but nontrivial to compute those of the union. In this paper, we show that if the original families are planar then the planar obstructions of the union are no larger than nO(n2), where n is the size of the largest obstruction of the original families.

The proof of Wagner's conjecture by N. Robertson and P. Seymour gives a finite description of any family of graphs which is closed under the minor ordering, called the obstructions of the family. Since the intersection and the union of two minor closed graph families are again a minor closed graph family, an interesting question is that of computing the obstructions of the new family given the obstructions for the original two families. It is easy to compute the obstructions of the intersection, but, until very recently, it was an open problem to compute the obstructions of the union. It is shown that if the original families are planar, then the obstructions of the union are no larger than n to the O(n/sup 2/) power, where n is the size of the largest obstruction of the original family.>