Current status of BioSmalltalk.

20 views
Skip to first unread message

wlcleve...@gmail.com

unread,
Jun 14, 2017, 9:59:05 PM6/14/17
to BioSmalltalk Users
Hello Herman and others,

It is great to see a Smalltalk-based tool for DNA analysis. Some years back, I wrote some code for a laboratory robot with VisualWorks and grew to really like the elegance of the Smalltalk language. Currently, I am just starting to analyze genomes in relation to psychiatric disorders. I just got my first family trio whole genome sequences. I am wondering if BioSmalltalk works well enough to be used as a practical tool and if many are using it. Would you recommend it for someone who has  a general familiarity with Smalltalk but has not used Pharo extensively and who is also a beginner in the DNA problem domain?

With much appreciation,

Lou Cleveland

Hernán Morales Durand

unread,
Jun 15, 2017, 1:27:45 AM6/15/17
to biosmallt...@googlegroups.com
Hello Lou,

Thank you for your interest. Here, in the Institute of Veterinary Genetics in Argentina, we are using BioSmalltalk since a few years. Most of my workflows are related with population genetics and phylogeography. Specifically I've used BioSmalltalk recently to interface PLINK 1.9, to generate input files and plot output for LAMP-LD (local ancestry inference), to run ShapeIt and PHASE (haplotype inference), generate code for rehh (detection of footprints of selection) and PhyloclassTalk for determining haplogroups using mtDNA geo-annotated sequences. NGS is a domain which I started to work actively since one year ago, but more applied to RNASeq than WGS. Right now, I am designing an object model to compare outputs from several trimmers, aligners and gene expression packages.

I would recommend BioSmalltalk to anyone who want more than copy and paste of bioinformatics snippets or commands. Definitely it is the ideal choice for someone who already knows Smalltalk and is learning sequence analysis, or want to model complex domains which are not very well explored or researched. Although other Bio* libraries currently seem to have more classes, too often is *very* difficult to evolve their code, i.e. to change something to do more than the documented examples.

About documentation. Yesterday I've uploaded a PDF with updated BioSmalltalk documentation:  https://github.com/hernanmd/web/raw/master/doc/BioSmalltalk-1.0.pdf . Although there are still many undocumented classes such as formatters for Browser Extensible Data (BED), MEGA, VCF (work in progress), etc.

Currently, I am trying to get an installer with more features, but in the meantime to install latest developement of BioSmalltalk, just open Pharo 5 and evaluate (tested on Windows 8.1):

Metacello new
    smalltalkhubUser: 'hernan' project: 'BioSmalltalk';
    configuration: 'BioSmalltalk';
    version: #bleedingEdge;
    load

Let me know if you have specific requirements or analysis in mind.

Best regards,

Hernán


--
You received this message because you are subscribed to the Google Groups "BioSmalltalk Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to biosmalltalk-users+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply all
Reply to author
Forward
0 new messages