Compare chainer example with GPU and without GPU.

Some days ago, @fmkz__ -san posted about chainer. http://blog.kzfmix.com/entry/1450445178 I have not ran example code, so tried it. At first, Run with out GPU. Next, Run with GPU. Only add option ‘–gpu=0’. 3 to 5 times faster with GPU than without GPU.

Hit triage, case of MTase

I often see keyword “PAINS” in literature . Frequent hitters of HTS deck are problematic and PAINS filter is useful to remove them because false positives are time and cost consuming. Researcher of Lilly, reported nice letter in J. Med. Chem. Lett. http://pubs.acs.org/doi/abs/10.1021/acsmedchemlett.5b00375 The author analysed there HTS campaign of Methyltransferases( MTase ). They screened

Make Docker image File with Chemoinformatics toolkits.

Docker https://www.docker.com is an open platform for distributed applications for developers and sysadmins. Providing docker image that means every one build same environment easily. It’s means create own image and share image. Docker provide container. Containers running on a single machine all share the same OS kernel. It’s difference of normal VM. I used docker

Set up GPU for chainer in OSx El Capitan.

I enjoyed a UGM that held on Nov. 5, 6 and hope participants also enjoyed the meeting. One of the topics of the meeting was “Deep Learning”(???). I introduced about python library for deep learning called ‘chainer’ . I presented my challenge to build model for QSAR, using chainer but results were ….(Hahaha) Chainer is

fish for drug discovery

Some years ago, I was interested in phenotypic screening that use zebrafish. In drug discovery project, we use animal models (often use rodent) to estimate efficacy, safety or etc. Also Knock out/in mouse is used for target validation. It’s still high cost and low throughput step. So, I think zebrafish is one of the interesting

make 3d PCA plot

I often use PCA(principal component analysis) to reduce dimension. I do PCA using Python sklearn or R language. Basic function of R “biplot” makes 2D chart. It’s easy way to make biplot. Today I found cool library of R, named “pca3d”. Install is easy! Just type following command. Now make chart. I used iris data

Metrics for Drug likeness

There are some indexes about drug likeness. Of course, I sometime use LE, LLE for lead optimization. Recently QED is new metrics for estimation of drug likeness. Which parameters do you like ? I read following paper that reported Patrick Barton et. al. It’s interesting for me. http://www.sciencedirect.com/science/article/pii/S1359644615003505 The author proposed new index named “AEI”

QSAR with chainer.

I wrote blog post about DL before. https://iwatobipen.wordpress.com/2014/07/07/deep-learning-with-python/ I used nolearn for QSAR in the post. There are some python library for DL. You know… Theano, nolearn, pylearn2, etc.. And chainer is new python library for neural networks. If reader interested the library, please search chainer in google, you will find lots of site or

Array or sparse array ?

In a process of lead optimization, chemist often do SAR expansion around potent compound. If lead compound can be break down three parts A(head), B(core), C(tail), chemist(me…) often fix one part(e.g core B) and change two parts. After optimize A and C then, fix A, C and change B. This approach is called array synthesis

Visualizing the process of lead optimization

Some time we set milestones to management of portfolio, or/and to check the progeress of projects. These data were reported document, power point slides etc, so it’s difficult to grasp situation of LO timely. Researchers at GSK published a solution of visualize LO process. It was impressive for me. Link is here. http://www.ncbi.nlm.nih.gov/pubmed/26262898 They called

Draw Molecular Matched Pair as SVG.

Somedays ago, I posted drawing molecule as SVG using RDKit. It works fine. So, I challenged draw MMP as SVG. My plan is … 1. Generate MMP using RDKit. 2. Store MMP data to MongoDB. 3. Provide MMP data to user using flask. 4. Draw structure on the fly using Ajax. OK, Let start! Step

Visualise dataset using seaborn.

I often use scatter plot to analyse relationship of 2 valuables. There are a lots of tools for visualise dataset. Now, I challenged to use seaborn. Seaborn is python library based on matplotlib. And easy to make cool visualisation. Following snippet, I read sample data from rdkit install folder and plot MolWt vs LogP using

Passport for compound.

I was interested in the title. “Compound Passport Service” http://dx.doi.org/10.1016/j.drudis.2015.06.011 AZ made passport for compound to manage compound rights tracking. The system can manage status of compounds, like ownership, permission and structure shared. I really impressed with the concept and system because I think that management of compound(and right) logistics is key factor in Drug

Draw molecule in RDKit

I really enjoyed OETK training. Thanks a lot y_sama and organiser of the training. And I told about rdkit with the TK specialist in break time. TK specialist asked me that can RDKit draw molecule with highlight substructure. I answered to him ‘yes… maybe…but I can’t show you example now…’. Now I searched Web and

repo up dateリポジトリ更新

I updated code about mishimasyk6. URL is following. https://github.com/iwatobipen/mishimasyk/tree/master/mishimasyk6_1 repo is https://github.com/iwatobipen/mishimasyk I added tooltip in the scatterplot. So, when user mouse over the point, tooltip will draw structure and mw, fr_sp3. D3js is powerful library for drawing charts. 勉強会でツールチップ実装できなかったって話したんで実装しました。 mishimasyl6_1のコードでは散布図を書いたあとマウスオーバーすると構造とかそれ以外のアノテーションを描画します。 d3.events.pageXでダイレクトにマウスの位置を検出してdivタグ要素にSVGを突っ込めば完了です。 ダイナミックな図ができて面白いと思います。 コードはこんな感じ もしレポジトリをDLして動かない場合はvar RDKit = rdk()のコメント部分を有効にすると動くと思います。 // var RDKit = rdk(); var padding

calculate 3d pharmacophore fingerprint in rdkit

I read document about RDKit for the first time in a while and found nice snippet. A 3D Pharmacophore finger print can be caluclated very easily. If dMat = None( default setting), fp means 2D Pharamcophore fingerprint, but now I passed dMat so the fingerprint has 3D information. Now 3D pharmacophore fingerprint in hand, it

Is CyTOF new technology ?

In some projects, pharmacologists showed me FACS data set. It’s useful to analyse cell response. I have not measured FACS by myself (I’m chemist…), so I didn’t understand details. Ala F. Nassar et al. published exciting review ( for me ) in DDT. About ‘mass’ cytometry. The link is following. http://www.sciencedirect.com/science/article/pii/S1359644615002317 I searched CyTOF as

Think about SAR analysis.

I lost a chance of participation in RDKIT-UGM because ticket was sold out. ;-( I’ll try next year…. SAR analysis is key for drug discovery. MMPA is one of major tool, I like the method. Because MMPA is easy to check effect of substituent in molecule. But sometime, it difficult to understand why the parameter

understand of ligand binding

I read the feature in this weekend. It’s useful information for me. Thermodynamics is very important but sometime I can’t use that for drug design. The review described some example about ligand binding thermodynamics in drug discovery. The most interested part for me was the story of thermodynamic profiling of BACE1 inhibitors. The author measured

Pharmacophore based molecular alignment

Sometime pharmacophore based alignment is used for scaffold hopping, virtual screening or etc. There are lots of famous tools to do that. Align-it is one of nice tool for mol align I think. It is open source tool. You can get the tool from silicos-it web site. http://silicos-it.be.s3-website-eu-west-1.amazonaws.com/index.html If reader who use OSx, align-it can