[FRIAM] Any non-biological complex systems?

[Marcus Daniels]

To clarify, one could contrast a procedure that resulted in the most reliable predictions but was not built on any ontology vs. one that could be communicated in a compact way and generalized to make other kinds of predictions but all of its predictions were not as reliable.

-----Original Message-----
From: Friam [mailto:friam-bounces at redfish.com] On Behalf Of Marcus Daniels
Sent: Tuesday, June 06, 2017 4:16 PM
To: The Friday Morning Applied Complexity Coffee Group <friam at redfish.com>
Subject: Re: [FRIAM] Any non-biological complex systems?

If one had full genome sequences for a lot of people and used a supervised learning procedure to predict the intellectual disability, that would not be result in a coherent explanation? 

-----Original Message-----
From: Friam [mailto:friam-bounces at redfish.com] On Behalf Of glen ep ropella
Sent: Tuesday, June 06, 2017 4:00 PM
To: The Friday Morning Applied Complexity Coffee Group <friam at redfish.com>
Subject: Re: [FRIAM] Any non-biological complex systems?

E.g. this seems like a coherent definition of coherence.  Is it lost in its technical detail?  Maybe.  Ontologies are rife with assumptions, like any other model.

Systematic Phenomics Analysis Deconvolutes Genes Mutated in Intellectual Disability into Biologically Coherent Modules
http://www.sciencedirect.com/science/article/pii/S0002929715004954

> Cluster biological coherence was calculated as follows. First, we retrieved the GO terms associated with the disease genes under- lying cluster syndromes. These GO terms were then pooled across all genes causing the same syndrome, resulting in a set of GO terms annotated to the syndrome. To incorporate ontological relation- ships between GO terms into the comparison, we added all term ancestors to the GO term set, excluding the root terms for the three GO categories. This approach has been shown to work as well as more complicated approaches.31 For each syndrome pair, we deter- mined the GO term overlap between the two syndromes:
> 
>   S_p(i,j) = n(G_i ∩ G_j)/n(G_i ∪ G_j)              (1)
> 
> where S_p(i,j) is the pairwise GO term overlap score for diseases i and j, n is the number of GO terms meeting the specified criteria, and G_i and G_j are the sets of GO terms associated with diseases i and j, respectively. For each cluster, the mean pairwise overlap was used as the biological coherence score for that cluster:
> 
>           n
>    S_c =  Σ S_p(i,j)/n
>          i,j
> 
> where S_c is the genetic cohesiveness score for cluster c, S_p(i,j) is the GO overlap score for diseases i and j, and n is the number of disease pairs in the cluster. The mean biological coherence score across all clusters was used as the overall cluster biological coherence score for the database:
> 
>        n
>    S = Σ S_c/n
>        c
> 
> where S is the overall genetic cohesiveness score for the phenotype data set, S_c is the genetic cohesiveness score for cluster c, and n is the number of clusters in the phenotype data set.

On 06/06/2017 08:33 AM, ┣glen┫ wrote:
> Second, your "interact more closely with one another than they do with entities outside the set" is nothing but _closure_.  Or if I can infer from the lack of response to my broaching the term, we could use "coherence" or some other word.  And that means that your working definition is not naive.  It does rely on an intuition that many of us share.  But in order for you to know what you're talking about, you have to apply a bit more formality to that concept.

--
glen ep ropella ⊥ 971-280-5699

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove ============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove