Damianos P. Melidis

Mr. Damianos P. Melidis

Leibniz University Hannover, Hanover, Germany
orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
12 October 2020 in Nucleic Acids Research
Verified
Journal Article
Open access
ACArif Canakoglu
Arif Canakoglu
PPPietro Pinoli
Pietro Pinoli
Anna BernasconiAnna Bernasconi
Anna Bernasconi
TATommaso Alfonsi
Tommaso Alfonsi
DMDamianos P. Melidis
Damianos P. Melidis
SCStefano Ceri
Stefano Ceri
ViruSurf, available at http://gmql.eu/virusurf/, is a large public database of viral sequences and integrated and curated metadata from heterogeneous sources (RefSeq, GenBank, COG-UK and NMDC); it also exposes computed nucleotide and amino acid variants, called from original sequences. A GISAID-specific ViruSurf database, available at http://gmql.eu/virusurf_gisaid/, offers a subset of these functionalities. Given the current pandemic outbreak, SARS-CoV-2 data are collected from the four sources; but ViruSurf contains other virus species harmful to humans, including SARS-CoV, MERS-CoV, Ebola and Dengue. The database is centered on sequences, described from their biological, technological and organizational dimensions. In addition, the analytical dimension characterizes the sequence in terms of its annotations and variants. The web interface enables expressing complex search queries in a simple way; arbitrary search queries can freely combine conditions on attributes from the four dimensions, extracting the resulting sequences. Several example queries on the database confirm and possibly improve results from recent research papers; results can be recomputed over time and upon selected populations. Effective search over large and curated sequence data may enable faster responses to future threats that could arise from new viruses.

19 January 2021 in Algorithms
Verified
Journal Article
Open access
PDF file
DMDamianos P. Melidis
Damianos P. Melidis
WNWolfgang Nejdl
Wolfgang Nejdl
Predicting biological properties of unseen proteins is shown to be improved by the use of protein sequence embeddings. However, these sequence embeddings have the caveat that biological metadata do not exist for each amino acid, in order to measure the quality of each unique learned embedding vector separately. Therefore, current sequence embedding cannot be intrinsically evaluated on the degree of their captured biological information in a quantitative manner. We address this drawback by our approach, dom2vec, by learning vector representation for protein domains and not for each amino acid base, as biological metadata do exist for each domain separately. To perform a reliable quantitative intrinsic evaluation in terms of biology knowledge, we selected the metadata related to the most distinctive biological characteristics of a domain, which are its structure, enzymatic, and molecular function. Notably, dom2vec obtains an adequate level of performance in the intrinsic assessment—therefore, we can draw an analogy between the local linguistic features in natural languages and the domain structure and function information in domain architectures. Moreover, we demonstrate the dom2vec applicability on protein prediction tasks, by comparing it with state-of-the-art sequence embeddings in three downstream tasks. We show that dom2vec outperforms sequence embeddings for toxin and enzymatic function prediction and is comparable with sequence embeddings in cellular location prediction.

21 September 2022 in PLOS Computational Biology
Verified
Journal Article
Open access
PDF file
DMDamianos P. Melidis
Damianos P. Melidis
CLChristian Landgraf
Christian Landgraf
GSGunnar Schmidt
Gunnar Schmidt
ASAnja Schöner-Heinisch
Anja Schöner-Heinisch
SHSandra von Hardenberg
Sandra von Hardenberg
BABernd Auber
Bernd Auber
Since next-generation sequencing (NGS) has become widely available, large gene panels containing up to several hundred genes can be sequenced cost-efficiently. However, the interpretation of the often large numbers of sequence variants detected when using NGS is laborious, prone to errors and is often difficult to compare across laboratories. To overcome this challenge, the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP) have introduced standards and guidelines for the interpretation of sequencing variants. Additionally, disease-specific refinements have been developed that include accurate thresholds for many criteria, enabling highly automated processing. This is of particular interest for common but heterogeneous disorders such as hearing impairment. With more than 200 genes associated with hearing disorders, the manual inspection of possible causative variants is particularly difficult and time-consuming. To this end, we developed the open-source bioinformatics tool GenOtoScope, which automates the analysis of all ACMG/AMP criteria that can be assessed without further individual patient information or human curator investigation, including the refined loss of function criterion (“PVS1”). Two types of interfaces are provided: (i) a command line application to classify sequence variants in batches for a set of patients and (ii) a user-friendly website to classify single variants. We compared the performance of our tool with two other variant classification tools using two hearing loss data sets, which were manually annotated either by the ClinGen Hearing Loss Gene Curation Expert Panel or the diagnostics unit of our human genetics department. GenOtoScope achieved the best average accuracy and precision for both data sets. Compared to the second-best tool, GenOtoScope improved the accuracy metric by 25.75% and 4.57% and precision metric by 52.11% and 12.13% on the two data sets, respectively. The web interface is accessible via: http://genotoscope.mh-hannover.de:5000 and the command line interface via: https://github.com/damianosmel/GenOtoScope.

18 March 2020 in bioRxiv
Verified
Preprint
Open access
PDF file
DMDamianos P. Melidis
Damianos P. Melidis
BMBrandon Malone
Brandon Malone
WNWolfgang Nejdl
Wolfgang Nejdl
Motivation: Word embedding approaches have revolutionized Natural Language Processing NLP research. These approaches aim to map words to a low-dimensional vector space in which words with similar linguistic features are close in the vector space. These NLP approaches also preserve local linguistic features, such as analogy. Embedding-based approaches have also been developed for proteins. To date, such approaches treat amino acids as words, and proteins are treated as sentences of amino acids. These approaches have been evaluated either qualitatively, via visual inspection of the embedding space, or extrinsically, via performance on a downstream task. However, it is difficult to directly assess the intrinsic quality of the learned embeddings. Results: In this paper, we introduce dom2vec, an approach for learning protein domain embeddings. We also present four intrinsic evaluation strategies which directly assess the quality of protein domain embeddings. We leverage the hierarchy relationship of InterPro domains, known secondary structure classes, Enzyme Commission class information, and Gene Ontology annotations in these assessments. These evaluations allow us to assess the quality of learned embeddings independently of a particular downstream task. Importantly, allow us to draw an analog between the local linguistic features in nature languages and the domain structure and function information in domain architectures, thus providing data-driven insights into the context found in the language of domain architectures. We also show that dom2vec embeddings outperform, or are comparable with, state-of-the-art approaches on downstream tasks. Availability: The protein domain embeddings vectors and the entire code to reproduce the results are available at https://github.com/damianosmel/dom2vec . Contact: melidis@l3s.uni-hannover.de

28 August 2020 in Research Square
Verified
Preprint
Open access
PDF file
DMDamianos P. Melidis
Damianos P. Melidis
BMBrandon Malone
Brandon Malone
WNWolfgang Nejdl
Wolfgang Nejdl
Background: Word embedding approaches have revolutionized natural language processing (NLP) research. These approaches aim to map words to a low-dimensional vector space, in which words with similar linguistic features cluster together. Embedding-based methods have also been developed for proteins, where words are amino acids and sentences are proteins. The learned embeddings have been evaluated qualitatively, via visual inspection of the embedding space and extrinsically, via performance comparison on downstream protein prediction tasks. However, these sequence embeddings have the caveat that biological metadata do not exist for each amino acid, in order to measure the quality of each unique learned embedding vector. Results: Here, we present dom2vec, an approach for learning protein domain embeddings using word2vec on InterPro annotations. In contrast to sequence embeddings, biological metadata do exist for protein domains, related to each domain separately. Therefore, we present four intrinsic evaluation strategies to quantitatively assess the quality of the learned embedding space. To perform a reliable evaluation in terms of biology knowledge, we selected the metadata related to the most distinctive biological characteristics of domains. These are the structure, enzymatic and molecular function of a given domain. Notably, dom2vec obtains adequate level of performance in the intrinsic assessment, therefore we can draw an analogy between the local linguistic features in natural languages and the domain structure and function information in domain architectures. Moreover, we demonstrate the dom2vec applicability on protein prediction tasks, by comparing it with state-of-the-art sequence embeddings in three downstream tasks. We show that dom2vec outperform sequence embeddings for toxin and enzymatic function prediction and is comparable with sequence embeddings in cellular location prediction. Conclusions: We report that the application of word2vec on InterPro annotations produces domain embeddings with two significant advantages over sequence embeddings. First, each unique dom2vec vector can be quantitatively evaluated towards its available structure and function metadata. Second, the produced embeddings can outperform the sequence embeddings for a subset of downstream tasks. Overall, dom2vec embeddings are able to capture the most important biological properties of domains and surpass sequence embeddings for a subset of prediction tasks.

24 December 2021 in bioRxiv
Verified
Preprint
Open access
PDF file
DMDamianos P. Melidis
Damianos P. Melidis
CLChristian Landgraf
Christian Landgraf
GSGunnar Schmidt
Gunnar Schmidt
ASAnja Schoener-Heinisch
Anja Schoener-Heinisch
SHSandra von Hardenberg
Sandra von Hardenberg
BABernd Auber
Bernd Auber
Since next-generation sequencing (NGS) has become widely available, large gene panels containing up to several hundred genes can be sequenced cost-efficiently. However, the interpretation of the often large numbers of sequence variants detected when using NGS is laborious, prone to errors and often not comparable across laboratories. To overcome this challenge, the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP) introduced standards and guidelines for the interpretation of sequencing variants. Further gene- and disease-specific refinements regarding hereditary hearing loss have been developed since then. With more than 200 genes associated with hearing disorders, the manual inspection of possible causative variants is especially difficult and time consuming. We developed an open-source bioinformatics tool GenOtoScope , which automates all ACMG/AMP criteria that can be assessed without further individual patient information or human curator investigation, including the refined loss of function criterion (“PVS1”). Two types of interfaces are provided: (i) a command line application to classify sequence variants in batches for a set of patients and (ii) a user-friendly website to classify single variants. We compared the performance of our tool with two other variant classification tools using two hearing loss data sets, which were manually annotated either by the ClinGen Hearing Loss Gene Curation Expert Panel or the diagnostics unit of our human genetics department. GenOtoScope achieved the best average accuracy and precision for both data sets. Compared to the second-best tool, GenOtoScope improved accuracy metric by 25.75% and 4.57% and precision metric by 52.11% and 12.13% on the two data sets respectively. The web interface is freely accessible. The command line application along with all source code, documentation and example outputs can be found via the project GitHub page. Author summary: New high-throughput sequencing technologies can produce massive amounts of information and are utilized by laboratories to explain the often complex genetic aetiology of hereditary diseases. The most common sensory disease, hearing loss, is often hereditary and has a high impact on a patient’s every-day life. To use these sequencing technologies effectively, software tools were developed that can aid researchers interpreting genetic data by semi-automatically classifying the biologic (and thus potentially medical) impact of detected variants (the alterations of the patient’s genome compared to the human reference genome). The available genetic variant classification tools are either not designed specifically for the interpretation of variants detected in subjects with hearing loss or they do not allow researchers to use them for batch classification of all variants detected, e. g. in a study group. To address this drawback, we developed GenOtoScope , an open-source tool that automates the pathogenicity classification of variants potentially associated with congenital hearing loss. GenOtoScope can be applied for the automatic classification of all variants detected in a set of probands.