Maciej Wójcikowski

ORCID: 0000-0003-1722-6797
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Computational Drug Discovery Methods
  • Bioinformatics and Genomic Networks
  • Machine Learning in Materials Science
  • Protein Structure and Dynamics
  • Click Chemistry and Applications
  • MicroRNA in disease regulation
  • Microbial Natural Products and Biosynthesis
  • Library Science and Information Systems
  • Genetics, Bioinformatics, and Biomedical Research
  • Biochemical and Structural Characterization
  • Cancer-related molecular mechanisms research
  • Biosimilars and Bioanalytical Methods
  • Research Data Management Practices
  • RNA modifications and cancer

Institute of Biochemistry and Biophysics, Polish Academy of Sciences
2013-2019

Abstract Classical scoring functions have reached a plateau in their performance virtual screening and binding affinity prediction. Recently, machine-learning trained on protein-ligand complexes shown great promise small tailored studies. They also raised controversy, specifically concerning model overfitting applicability to novel targets. Here we provide new ready-to-use function (RF-Score-VS) 15 426 active 893 897 inactive molecules docked set of 102 We use the full DUD-E data sets along...

10.1038/srep46710 article EN cc-by Scientific Reports 2017-04-25

There has been huge progress in the open cheminformatics field both methods and software development. Unfortunately, there little effort to unite those into one package. We here describe Open Drug Discovery Toolkit (ODDT), which aims fulfill need for comprehensive source drug discovery software. The was developed as a free tool computer aided (CADD) developers researchers. ODDT reimplements many state-of-the-art methods, such machine learning scoring functions (RF-Score NNScore) wraps other...

10.1186/s13321-015-0078-2 article EN cc-by Journal of Cheminformatics 2015-06-11

Abstract Motivation Fingerprints (FPs) are the most common small molecule representation in cheminformatics. There a wide variety of FPs, and Extended Connectivity Fingerprint (ECFP) is one best-suited for general applications. Despite overall FP abundance, only few FPs represent 3D structure molecule, hardly any encode protein–ligand interactions. Results Here, we present Protein–Ligand (PLEC) that implicitly encodes interactions by pairing ECFP environments from ligand protein. PLEC were...

10.1093/bioinformatics/bty757 article EN cc-by-nc Bioinformatics 2018-09-06

Abstract Summary: MiRNAs are short, non-coding molecules that negatively regulate gene expression and thereby play several important roles in living organisms. Dozens of computational methods for miRNA-related research have been developed, which greatly differ various aspects. The substantial availability difficult-to-compare approaches makes it challenging the user to select a proper tool prompts need solution will collect categorize all methods. Here, we present tools4miRs, first platform...

10.1093/bioinformatics/btw189 article EN cc-by-nc Bioinformatics 2016-04-08

Fingerprints (FPs) are the most common small molecule representation in cheminformatics. There a wide variety of fingerprints, and Extended Connectivity Fingerprint (ECFP) is one best-suited for general applications. Despite overall FP abundance, only few FPs represent 3D structure molecule, hardly any encode protein-ligand interactions. Here, we present Protein-Ligand (PLEC) fingerprint that implicitly encodes interactions by pairing ECFP environments from ligand protein. PLEC fingerprints...

10.26434/chemrxiv.5928406.v1 preprint EN 2018-02-28

DiSCuS, a "Database System for Compound Selection", has been developed. The primary goal of DiSCuS is to aid researchers in the steps subsequent generating high-throughput virtual screening (HTVS) results, such as selection compounds further study, purchase, or synthesis. To do so, provides (1) storage facility ligand-receptor complexes (generated with external programs), (2) number tools validating these complexes, scoring functions, potential energy contributions, and med-chem features...

10.1021/ci400587f article EN publisher-specific-oa Journal of Chemical Information and Modeling 2013-12-23

<div>Fingerprints (FPs) are the most common small molecule representation in cheminformatics. There a wide variety of fingerprints, and Extended Connectivity Fingerprint (ECFP) is one best-suited for general applications. Despite overall FP abundance, only few FPs represent 3D structure molecule, hardly any encode protein-ligand interactions. Here, we present Protein-Ligand (PLEC) fingerprint that implicitly encodes interactions by pairing ECFP environments from ligand protein. PLEC...

10.26434/chemrxiv.5928406 preprint EN 2018-02-28
Coming Soon ...