Tomas Klingström

Protein–protein interaction and pathway databases, a graphical review

The amount of information regarding protein-protein interactions (PPI) at a proteomic scale is constantly increasing. This is paralleled with an increase of databases making information available. Consequently there are diverse ways of delivering information about not only PPIs but also regarding the databases themselves. This creates a time consuming obstacle for many researchers working in the field. Our survey provides a valuable tool for researchers to reduce the time necessary to gain a broad overview of PPI-databases and is supported by a graphical representation of data exchange. The graphical representation is made available in cooperation with the team maintaining www.pathguide.org and can be accessed at http://www.pathguide.org/interactions.php in a new Cytoscape web implementation. The local copy of Cytoscape cys file can be downloaded from http://bio.icm.edu.pl/~darman/ppi web page.

Ensemble learning prediction of protein–protein interactions using proteins functional annotations

Protein-protein interactions are important for the majority of biological processes. A significant number of computational methods have been developed to predict protein-protein interactions using protein sequence, structural and genomic data. Vast experimental data is publicly available on the Internet, but it is scattered across numerous databases. This fact motivated us to create and evaluate new high-throughput datasets of interacting proteins. We extracted interaction data from DIP, MINT, BioGRID and IntAct databases. Then we constructed descriptive features for machine learning purposes based on data from Gene Ontology and DOMINE. Thereafter, four well-established machine learning methods: Support Vector Machine, Random Forest, Decision Tree and Naïve Bayes, were used on these datasets to build an Ensemble Learning method based on majority voting. In cross-validation experiment, sensitivity exceeded 80% and classification/prediction accuracy reached 90% for the Ensemble Learning method. We extended the experiment to a bigger and more realistic dataset maintaining sensitivity over 70%. These results confirmed that our datasets are suitable for performing PPI prediction and Ensemble Learning method is well suited for this task. Both the processed PPI datasets and the software are available at .

Workshop on laboratory protocol standards for the molecular methods database

Management of data to produce scientific knowledge is a key challenge for biological research in the 21st century. Emerging high-throughput technologies allow life science researchers to produce big data at speeds and in amounts that were unthinkable just a few years ago. This places high demands on all aspects of the workflow: from data capture (including the experimental constraints of the experiment), analysis and preservation, to peer-reviewed publication of results. Failure to recognise the issues at each level can lead to serious conflicts and mistakes; research may then be compromised as a result of the publication of non-coherent protocols, or the misinterpretation of published data. In this report, we present the results from a workshop that was organised to create an ontological data-modelling framework for Laboratory Protocol Standards for the Molecular Methods Database (MolMeth). The workshop provided a set of short- and long-term goals for the MolMeth database, the most important being the decision to use the established EXACT description of biomedical ontologies as a starting point.

A comprehensive model of DNA fragmentation for the preservation of High Molecular Weight DNA

For long-read sequencing applications, shearing of DNA is a significant issue as it limits the read-lengths generated by sequencing. During extraction and storage of DNA the DNA polymers are susceptible to physical and chemical shearing. In particular, the mechanisms of physical shearing are poorly understood in most laboratories as they are of little relevance to commonly used short-read sequencing technologies. This study draws upon lessons learned in a diverse set of research fields to create a comprehensive theoretical framework for obtaining high molecular weight DNA (HMW-DNA) to support improved quality management in laboratories and biobanks for long-read sequencing applications. Under common laboratory conditions physical and chemical shearing yields DNA fragments of 5-35 kilobases (kb) in length. This fragment length is sufficient for DNA sequencing using short-read technologies but for Nanopore sequencing, linked reads and single molecular real time sequencing (SMRT) poorly preserved DNA will limit the length of the reads generated. The shearing process can be divided into physical and chemical shearing which generates different patterns of fragmentation. Exposure to physical shearing creates a characteristic fragment length where the main cause of shearing is shear stress induced by turbulence. The characteristic fragment length is several thousand base pairs longer than the reads produced by short-read sequencing as the shear stress imposed on short DNA fragments is insufficient to shear the DNA. This characteristic length can be measured using gel electrophoresis or instruments for DNA fragment analysis,. Chemical shearing generates randomly distributed fragment lengths visible as a smear of DNA below the peak fragment length. By measuring the peak of the DNA fragment length distribution and the proportion of very short DNA fragments, both sources of shearing can be measured using commonly used laboratory techniques, providing a suitable quantification of DNA integrity of DNA for sequencing with long-read technologies.

Legal & ethical compliance when sharing biospecimen

Abstract When obtaining samples from biobanks, resolving ethical and legal concerns is a time-consuming task where researchers need to balance the needs of privacy, trust and scientific progress. The Biobanking and Biomolecular Resources Research Infrastructure-Large Prospective Cohorts project has resolved numerous such issues through intense communication between involved researchers and experts in its mission to unite large prospective study sets in Europe. To facilitate efficient communication, it is useful for nonexperts to have an at least basic understanding of the regulatory system for managing biological samples. Laws regulating research oversight are based on national law and normally share core principles founded on international charters. In interview studies among donors, chief concerns are privacy, efficient sample utilization and access to information generated from their samples. Despite a lack of clear evidence regarding which concern takes precedence, scientific as well as public discourse has largely focused on privacy concerns and the right of donors to control the usage of their samples. It is therefore important to proactively deal with ethical and legal issues to avoid complications that delay or prevent samples from being accessed. To help biobank professionals avoid making unnecessary mistakes, we have developed this basic primer covering the relationship between ethics and law, the concept of informed consent and considerations for returning findings to donors.

The eBioKit, a stand-alone educational platform for bioinformatics

Bioinformatics skills have become essential for many research areas; however, the availability of qualified researchers is usually lower than the demand and training to increase the number of able bioinformaticians is an important task for the bioinformatics community. When conducting training or hands-on tutorials, the lack of control over the analysis tools and repositories often results in undesirable situations during training, as unavailable online tools or version conflicts may delay, complicate, or even prevent the successful completion of a training event. The eBioKit is a stand-alone educational platform that hosts numerous tools and databases for bioinformatics research and allows training to take place in a controlled environment. A key advantage of the eBioKit over other existing teaching solutions is that all the required software and databases are locally installed on the system, significantly reducing the dependence on the internet. Furthermore, the architecture of the eBioKit has demonstrated itself to be an excellent balance between portability and performance, not only making the eBioKit an exceptional educational tool but also providing small research groups with a platform to incorporate bioinformatics analysis in their research. As a result, the eBioKit has formed an integral part of training and research performed by a wide variety of universities and organizations such as the Pan African Bioinformatics Network (H3ABioNet) as part of the initiative Human Heredity and Health in Africa (H3Africa), the Southern Africa Network for Biosciences (SAnBio) initiative, the Biosciences eastern and central Africa (BecA) hub, and the International Glossina Genome Initiative.

Supporting the development of biobanks in low and medium income countries

Biobanks are an organized collection of biological material and associated data. They are a fundamental resource for life science research and contribute to the development of pharmaceutical drugs, diagnostic markers and to a deeper understanding of the genetics that regulate the development of all life on earth. Biobanks are well established in High Income Countries (HIC) and are rapidly emerging in Low and Middle Income Countries (LMIC). Surveys among biobanks operating in a LMIC setting indicate that limited resources and short term funding tied to specific projects threaten the sustainability of the biobanks. Fit-for-purpose biobanks targeting major societal challenges such as HIV and Malaria provide an excellent basis for integrating biobanks with the available research communities in LMIC regions. But to become sustainable for the future it is important that biobanks become an integrated part of local research communities. To achieve this, the cost of operating biobanks must be lowered, templates must be developed to support local ethics committees and researchers must be given the opportunity to build experience in successfully operating biobank based research projects. The B3Africa consortium is based on these conclusions and set up to support biobank based research by creating a cost efficient Laboratory Information Management System (LIMS) for developing biobanks and also contribute to the training and capacity building in the local research community. The technical platform called the eB3Kit is open source and consists of a LIMS and a bioinformatics module based on the eBiokit that allow researchers to take control over the analysis of their own data. Along with the technical platform the consortium will also contribute training and support for the associated infrastructures necessary to regulate the ethical and legal implications of biobank based research.