Chemoinformatics and structural bioinformatics in OCaml

Berenger, Francois; Zhang, Kam Y. J.; Yamanishi, Yoshihiro

doi:10.1186/s13321-019-0332-0

Cited by 5 publications

(7 citation statements)

References 60 publications

(54 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…OCaml is a strongly typed functional language . Types are automatically inferred and enforced at compile-time, removing several classes of run-time errors . Our molecular encoder is also released as open source .…”

Section: Methodsmentioning

confidence: 92%

See 1 more Smart Citation

Ranking Molecules with Vanishing Kernels and a Single Parameter: Active Applicability DomainIncluded

Berenger

Yamanishi

2020

J. Chem. Inf. Model.

Self Cite

View full text Add to dashboard Cite

In ligand-based virtual screening, high-throughput screening (HTS) data sets can be exploited to train classification models. Such models can be used to prioritize yet untested molecules, from the most likely active (against a protein target of interest) to the least likely active. In this study, a single-parameter ranking method with an Applicability Domain (AD) is proposed. In effect, Kernel Density Estimates (KDE) are revisited to improve their computational efficiency and incorporate an AD. Two modifications are proposed: (i) using vanishing kernels (i.e., kernel functions with a finite support) and (ii) using the Tanimoto distance between molecular fingerprints as a radial basis function. This construction is termed "Vanishing Ranking Kernels" (VRK). Using VRK on 21 HTS assays, it is shown that VRK can compete in performance with a graph convolutional deep neural network. VRK are conceptually simple and fast to train. During training, they require optimizing a single parameter. A trained VRK model usually defines an active AD.

show abstract

Section: Methodsmentioning

confidence: 92%

“…70 Types are automatically inferred and enforced at compile-time, removing several classes of run-time errors. 71 Our molecular encoder is also released as open source. 43 On multicore computers, RanKers is parallelized with the Parmap library.…”

Section: ■ Methodsmentioning

confidence: 99%

Ranking Molecules with Vanishing Kernels and a Single Parameter: Active Applicability DomainIncluded

Berenger

Yamanishi

2020

J. Chem. Inf. Model.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Software implementation Typing atoms and bonds, selecting and tagging cleaved bonds is done by a Python script using RDKit [ 38 ]. For performance and correctness reasons [ 52 ], indexing fragments and fragment assembly is done by an OCaml [ 53 ] program reading SMILES with tagged cleaved bonds. The program is named FASMIFRA for “Fast Assembly of SMILES Fragments”.…”

Section: Methodsmentioning

confidence: 99%

Molecular generation by Fast Assembly of (Deep)SMILES fragments

Berenger

Tsuda

2021

J Cheminform

Self Cite

View full text Add to dashboard Cite

Background In recent years, in silico molecular design is regaining interest. To generate on a computer molecules with optimized properties, scoring functions can be coupled with a molecular generator to design novel molecules with a desired property profile. Results In this article, a simple method is described to generate only valid molecules at high frequency ($$>300,000$$ > 300 , 000 molecule/s using a single CPU core), given a molecular training set. The proposed method generates diverse SMILES (or DeepSMILES) encoded molecules while also showing some propensity at training set distribution matching. When working with DeepSMILES, the method reaches peak performance ($$>340,000$$ > 340 , 000 molecule/s) because it relies almost exclusively on string operations. The “Fast Assembly of SMILES Fragments” software is released as open-source at https://github.com/UnixJunkie/FASMIFRA. Experiments regarding speed, training set distribution matching, molecular diversity and benchmark against several other methods are also shown.

show abstract

“…In this thematic series we have invited authors to present their views on a variety of programming languages. The series is rolling, and starts of with contributions from Thiesen [16], Berenger [17], and Höck [18] discussing JavaScript, OCaml and Scala respectively. We anticipate contributions covering Scala, C/C++, Tcl and noSQL.…”

Section: Contentmentioning

confidence: 99%

Implementing cheminformatics

Guha

2019

J Cheminform

View full text Add to dashboard Cite

Chemoinformatics and structural bioinformatics in OCaml

Cited by 5 publications

References 60 publications

Ranking Molecules with Vanishing Kernels and a Single Parameter: Active Applicability DomainIncluded

Ranking Molecules with Vanishing Kernels and a Single Parameter: Active Applicability DomainIncluded

Molecular generation by Fast Assembly of (Deep)SMILES fragments

Implementing cheminformatics

Contact Info

Product

Resources

About