Martin Krzywinski / Genome Sciences Center / mkweb.bcgsc.ca Martin Krzywinski / Genome Sciences Center / mkweb.bcgsc.ca - contact me Martin Krzywinski / Genome Sciences Center / mkweb.bcgsc.ca on Twitter Martin Krzywinski / Genome Sciences Center / mkweb.bcgsc.ca - Lumondo Photography Martin Krzywinski / Genome Sciences Center / mkweb.bcgsc.ca - Pi Art Martin Krzywinski / Genome Sciences Center / mkweb.bcgsc.ca - Hilbertonians - Creatures on the Hilbert Curve
Twenty — minutes — maybe — more.Naomichoose four wordsmore quotes

words: fun


In Silico Flurries: Computing a world of snow. Scientific American. 23 December 2017


language + fiction

Dark Matter of the English Language—the unwords

Words are easy, like the wind;
Faithful friends are hard to find.
—William Shakespeare

The unwords are words that are not in a language. Here I provide lists of such words, generated using a neural network.

Who knows? You may be the first to read these newly minted words and, armed with this knowlege, you can finally manufacture that exotic malady you've been looking for to help you get out of (or into) a party.

If you're a writer, these lists are a great resource for fictitious character names, places, animals, and cars.

I have a separate page of neologisms that I have created.

generating new words, names and places

The word lists were generated using the char-rnn recurrent neural network. Depending on the list the inputs were a large English dictionary, first names, names of countries or drug names.

This process is fun and the output is often believably hilarious. The application of char-rnn to the creation of new paint names, such as turdly and Ronching blue, is what motivated me to try this out for myself. Ronch on!

So, if you want to explore Conchar and Pobacia and Sinistan or wonder what kind of symptoms are associated with myconomascophobia or are curious about what a cakmiran would look like, explore the lists in depth and use a new word tonight.

Previous efforts

I've previously used Markov Chains to generate Tripsum, random ramblings of the mad Donald Trump, or Hitchum, the random ramblings of the brilliant Christopher Hitchens.

Quick Examples

Here are a few short lists to get you started. They're all sampled from my list of about 883,500 English unwords.

fields of study

Absolutely perfect for that time you don't really want to tell someone what you do. Shut down a conversation quickly with necronology or start one with jabimanology.

—7—
agology
beology
deology
egology
feology
giology
idology
phology
teology
viology
—8—
addology
belology
cagology
damology
ecpology
fenology
gadology
hedology
ingology
macology
nezology
oppology
pacology
senology
tanology
undology
—9—
acceology
bectology
cantology
daniology
egotology
fanzology
gaenology
heltology
icerology
lellology
maerology
neevology
ochrology
palvology
regmology
sentology
taltology
uricology
veleology
—10—
acctiology
bebolology
cadenology
decheology
ecotiology
faranology
galfiology
hegosology
ibberology
kerimology
lekanology
madorology
nedanology
odidiology
pacreology
regonology
saconology
tanzoology
unfelology
verevology
warryology
—11—
acestnology
baomasology
cacarrology
damakhology
eabidiology
fammocology
gactomology
icedorology
kannesology
leamonology
magoulology
necronology
orchitology
parcimology
rassamology
spetitology
talmatology
untectology
vertulology
wireulology
zarcidology
—12—
aderentology
bailiomology
cacometology
daiomatology
ecyphenology
finaginology
gaarozoology
hematizology
icernoxology
jabimanology
kenamidology
macopicology
noidentology
octuditology
palamanology
recestiology
spactomology
tamanthology
velarivology
—13—
acctohicology
beonephnology
caconyonology
damalanuology
ecerchenology
gecheepsology
helopatrology
idolotamology
kiliagorology
lamolatrology
madichonology
nonemartology
omphonomology
perculonology
seperahiology
tepromorology
unsophenology
velarithology

trump's best words

More tremendous than before, here is a sneak peek at the President's new vocabulary. He was recently seen acting reassoritious, though aides worry about his gabalkerous tendencies.

—7—
acanous
bateous
cabrous
dachous
ecelous
fallous
gabuous
harious
iberous
kangous
lelious
maceous
nagious
obbious
pahyous
querous
ratwous
scatous
taimous
utugous
veysous
wirbous
zabrous
—8—
abellous
bacerous
caborous
dachious
eablious
facomous
gaarious
havamous
ibbicous
jabimous
kamulous
lamanous
mabilous
namamous
octerous
pacorous
quawnous
raralous
sigurous
tacheous
ufturous
vhodious
weybrous
yepenous
—9—
abrybrous
baltylous
cabellous
dactinous
eccarious
facuelous
gadacrous
harrazous
icabulous
jankarous
kantalous
lammanous
mabecious
nabocious
oadaceous
pacculous
quebunous
ramperous
sacculous
tacallous
udnygnous
vebageous
whioteous
yiembrous
—10—
ablepelous
bailligous
cabalizous
dabsterous
eacopagous
fadiettous
gacaphuous
hasikerous
iabiaceous
jabimarous
kaguicgous
lamphylous
macaritous
naantalous
ocobarious
pacolagous
quacidious
rardaceous
saconogous
tabitolous
uebbaceous
veeaterous
waniferous
yridyovous
—11—
abbuliglous
bailouheous
caccantrous
dabsiparous
ebamicarous
falaealpous
gabalkerous
harspillous
icacalcious
jenkatapous
kedamatrous
lacluncrous
maanigerous
nabitarious
oacreparous
pactavinous
quedyismous
reangrevous
sabriconous
tacopathous
ueduraceous
velelaucous
warniferous
yupliferous
—12—
abrybraceous
bailligirous
cabstiferous
dabnasaceous
eabycerinous
fanatombrous
gactitarious
harsperstous
icaneopylous
jacugnaceous
kabridiflous
lanundineous
maagilineous
nacocarphous
oachophagous
pacanigenous
rambreaulous
sackniferous
tachyphonous
ulonollurous
vacoraageous
weloniaceous
zyanisudious
—13—
absiaphaceous
badaldiaceous
cabolophagous
dachariaceous
eccaralculous
facoschaceous
gacultiaceous
helaptoconous
icancraperous
jejuctogenous
karaseerheous
leaageistrous
macalopathous
namanocoalous
oacantraceous
pabliporagous
quaculiferous
reassoritious
saconophorous
tacalasmatous
unceanigerous
vegnuriferous

get thee to a fefery!

Building on covfefe, here are the words containing fefe that the neural network discovered, without covfefe being in the training set.

*fefe*: befefeeper brefefent brefefer cofefesce fefee fefeing fefenially fefenialness fefenian fefenianity fefeniate fefeoridy feferate feferated feferating feferats feferial feferiality feferially feferialness feferic feferies feferonious feferoptable fefery fefes gafefeyed gufefer hifefeed tifeferreuse

personal values

What do you hold dearest to your heart? You may not want to answer that question but you don't want to say nothing. Assess your moral superiority by loudly professing your views on abobionalism.

—6—
ablism
banism
cacism
defism
echism
famism
gadism
hihism
ichism
katism
lipism
madism
nadism
oacism
piaism
savism
tatism
udsism
vhrism
wipism
—7—
abatism
baggism
cabeism
dabyism
ecelism
famyism
gaatism
hekkism
icerism
kaggism
leliism
maatism
oinyism
paarism
redrism
sabyism
tagsism
vettism
wannism
zyarism
—8—
abxecism
bachrism
cabitism
dailyism
eacoxism
falerism
gacolism
haufoism
icastism
jabicism
kecarism
ladedism
mabreism
naimdism
oinicism
pakerism
quedyism
ramblism
saxicism
tadalism
uejogism
vavbeism
warryism
zerryism
—9—
abamerism
bacoinism
cablacism
dacantism
ecarcuism
falgonism
gaabilism
hafletism
icemagism
jabergism
kanmitism
lambinism
mackelism
nammalism
oabateism
pacoleism
ramemaism
saboirism
tacialism
uejoinism
velingism
werialism
—10—
aboboidism
basozacism
cablionism
daffialism
eccynatism
facomanism
gaariotism
helgralism
icastivism
jabicalism
kadharoism
lalaippism
maagillism
nadenicism
oadantyism
pacodotism
ramanguism
sacagenism
tabuxalism
uemitalism
vefflenism
whindanism
—11—
abariiarism
barkepolism
cabelierism
dabsolitism
eactubilism
falgmantism
gacticalism
harrigonism
icanepotism
jakekannism
kanemortism
lacloparism
mabinianism
nairsianism
oachanokism
pacanialism
rechapolism
secolastism
tacdicalism
unmantsmism
venianchism
weistentism
zabinianism
—12—
abobionalism
balfationism
caboonialism
dabnationism
eacesicalism
famiticanism
gabrinantism
hedorelicism
icasyntheism
kanmiphagism
lelipetylism
mackbelloism
nalasmandism
omphonophism
paargomelism
quadillotism
reamoplagism
scuntylinism
tachyphonism
uequitionism
vabippareism
wircationism

who are these people?

It's hard to know who you are. And even harder admitting it in public. Pick something safe from this list of identities. Be the gabdologist you've always wanted.

—6—
accist
bebist
cacist
dakist
egyist
farist
gadist
hadist
ichist
jamist
kawist
madist
nlhist
oacist
padist
revist
samist
tawist
uelist
vidist
wgeist
—7—
abolist
baggist
cabrist
dabvist
ebalist
falgist
gaboist
heddist
iccoist
jabrist
kaggist
leltist
maclist
neerist
olanist
palpist
redoist
sogyist
tabuist
uejoist
vequist
wannist
zaivist
—8—
ablarist
bachrist
cacknist
dabsaist
ecercist
faletist
gachlist
haufoist
iccomist
jacklist
kalarist
lepanist
maculist
nenalist
odiarist
paintist
reashist
saxinist
tadalist
uejogist
velewist
wandaist
—9—
acarenist
baelikist
cabourist
dactinist
ecgartist
falgonist
gadantist
hcmedrist
icabilist
jelimyist
kaggitist
lammetist
mackemist
nacldoist
oberalist
pablinist
quahhoist
rastarist
sabrilist
tablanist
udnygnist
veestoist
wellhoist
—10—
abcrappist
badaladist
cacherlist
dabuschist
eccularist
facomanist
gachtraist
haddootist
ibbicarist
jalatalist
kalanalist
ladulloist
maccrecist
najumanist
oacologist
pacolikist
recharmist
sabamagist
talmangist
uemitalist
vegonexist
whelentist
—11—
acceologist
baiskallist
cabodianist
daffudalist
eccbietrist
fanationist
gabdologist
harrigonist
icefomalist
kedamietist
lebsynsmist
maarheadist
naniquinist
ocpenvalist
pacanialist
recharmwist
seckneolist
tacdicalist
uncherylist
velelaucist
wenmologist
zyanisthist
—12—
abutationist
becorrhygist
cacomelanist
dabnationist
eacesicalist
fanineralist
gactionalist
hedephardist
ibaidalomist
kaththedrist
lanronatrist
macalomanist
nalasmandist
oinylandrist
pacreologist
quahentalist
reamoplagist
scatometrist
tamalogicist
undongiodist
vaxgallowist
whaslugnoist

time for action

So, what are your plans for the future? If you're one of those who feel the question is unanswerable with the current option of words, consider fallupify as a course of action. And if you're feeling less productive but still want to appear like you're doing something, go ahead and nonerify.

—6—
abdify
baeify
cafify
dalify
endify
falify
galify
havify
ichify
kerify
leaify
maeify
nozify
oodify
palify
rasify
saxify
talify
untify
velify
—7—
afffify
bektify
cackify
dankify
eddeify
fandify
gachify
haurify
icacify
jympify
kersify
maccify
necrify
orchify
pelnify
quacify
reanify
seerify
tactify
vettify
whizify
—8—
admanify
beennify
cacktify
daishify
encutify
falutify
ganprify
hemarify
inchoify
jadinify
keethify
ledidify
maddrify
nonerify
oliquify
paltrify
reassify
sebanify
tavetify
unducify
veridify
wirssify
—9—
bedaunify
caesonify
debissify
ecquilify
fallupify
gaaggrify
helantify
idrobrify
lacrusify
magvalify
omphonify
panratify
rejursify
sedfigify
tabjacify
uncepnify
wipencify
—10—
baltharify
caccretify
degiwhsify
enbificify
flodistify
gedlecrify
hevamilify
icertilify
leliponify
mannentify
nirgresify
oframarify
pencomnify
quadiasify
reflectify
temerdrify
uncmencify
wontistify
—11—
bemiflacify
calapaulify
deflestrify
ghakabilify
incandacify
manciculify
outsentrify
precrictify
rewerricify
thoullutify
ungargunify
—12—
anticulptify
bennebersify
cenorshanify
disommencify
eodyshotrify
michthillify
prirosaurify

no whimpers here, just bangs—here come the –omes

I've previously been reference as making fun —omes in the New York Times article "Ome," the Sound of the Scientific Universe Expanding.

Well, now I have even more —omes to ridicule. Hilariously, femome and manome are options!

—6—
adnome
agrome
balome
bdbome
cacome
canome
darome
debome
eccome
eedome
farome
femome
gamome
garome
hedome
horome
indome
infome
konome
lamome
madome
manome
octome
outome
pacome
palome
refome
theome
tibome
untome
venome
—7—
adesome
adonome
battome
bectome
callome
cannome
decrome
dedrome
ecthome
edihome
fammome
faysome
gaddome
gadrome
hargome
hidbome
icisome
inclome
jansome
kabrome
kawsome
lammome
maarome
madrome
nefiome
norcome
pagwome
pastome
rowcome
ruphome
sentome
soysome
tachome
taksome
unclome
veesome
vootome
weysome
—8—
abrosome
abzizome
becksome
becplome
caanrome
caconome
dacktome
dedesome
ecottome
eedotome
fandsome
farksome
gabarome
gabosome
heamsome
hedesome
ichanome
ichilome
kelatome
kelaxome
lailsome
lentsome
maarsome
machsome
nadidome
nalatome
oleinome
opeccome
paimsome
palesome
ratesome
reffrome
sefisome
sepreome
tacosome
tebanome
unchrome
voortome
wirssome
wiserome
—9—
acdeasome
acleusome
baudylome
beconsome
caliocome
caliosome
dactogome
dalatcome
eccistome
ecpostome
facoscome
fanersome
gaberrome
gaenesome
heezatome
helaptome
icesimome
iddicrome
kannisome
kauffsome
leneanome
londarome
mabersome
macrasome
necrotome
noenesome
orphecome
palmetome
paloktome
reciitome
retrotome
sepostome
slondmome
tarrocome
tealocome
udnygrome
undretome
veltosome
vewedsome
whealsome
whieltome
—10—
alonoptome
angiagrome
bejurgsome
belarizome
cabblesome
cabroorome
danamasome
decreotome
egieuctome
enoccotome
fanzoflome
fattstrome
gamomonome
ganadodome
hecalotome
helapptome
incanizome
inclostome
jabimatome
kainersome
karochrome
lelitoxome
linkledome
macenolome
mackahsome
nondyctome
nontentome
oholostome
omyrhysome
panapesome
pecoefsome
reclustome
regpensome
seerensome
segionsome
teadostome
teasodrome
unberwrome
unchlesome
verevisome
vhroustome
whimsesome
wirblosome
—11—
acctiphsome
acynotopome
bebrolysome
beeltectome
cabouthsome
calcomotome
dakesthrome
danjoirsome
ecpoplesome
eganulitome
fepelessome
fertostrome
gacreiovome
gactorrhome
hansboysome
hefrodesome
indaloctome
indanactome
lolaductome
mackmotsome
malaversome
nenacestome
noranectome
omynymysome
oppolionome
pacuderhome
paliplesome
risionfsome
spiguissome
stiltlesome
teentrotome
tenolustome
ungrefysome
unraxtchome
venianizome
vinacectome
wortarisome
—12—
acctiomyxome
actiprectome
basziversome
bebromyerome
calcoriotome
candochotome
deconnersome
decturessome
ecophytotome
egytherotome
fodbirdisome
fordharksome
gellardesome
gellmarksome
hernoelotome
hotchielsome
incyprossome
internantome
kalachiteome
keonylminome
lamphylotome
mackbanrsome
mafantectome
ochrophotome
octidraitome
pathymiotome
pecthenizome
rillpeypwome
rimentorrome
spotupersome
tarilocosome
thyrapioxome
undrarthrome

VIEW ALL

news + thoughts

Machine learning: supervised methods (SVM & kNN)

Thu 18-01-2018
Supervised learning algorithms extract general principles from observed examples guided by a specific prediction objective.

We examine two very common supervised machine learning methods: linear support vector machines (SVM) and k-nearest neighbors (kNN).

SVM is often less computationally demanding than kNN and is easier to interpret, but it can identify only a limited set of patterns. On the other hand, kNN can find very complex patterns, but its output is more challenging to interpret.

Martin Krzywinski @MKrzywinski mkweb.bcgsc.ca
Nature Methods Points of Significance column: Machine learning: supervised methods (SVM & kNN). (read)

We illustrate SVM using a data set in which points fall into two categories, which are separated in SVM by a straight line "margin". SVM can be tuned using a parameter that influences the width and location of the margin, permitting points to fall within the margin or on the wrong side of the margin. We then show how kNN relaxes explicit boundary definitions, such as the straight line in SVM, and how kNN too can be tuned to create more robust classification.

Bzdok, D., Krzywinski, M. & Altman, N. (2018) Points of Significance: Machine learning: a primer. Nature Methods 15:5–6.

Background reading

Bzdok, D., Krzywinski, M. & Altman, N. (2017) Points of Significance: Machine learning: a primer. Nature Methods 14:1119–1120.

...more about the Points of Significance column

Human Versus Machine

Tue 16-01-2018
Balancing subjective design with objective optimization.

In a Nature graphics blog article, I present my process behind designing the stark black-and-white Nature 10 cover.

Nature 10, 18 December 2017

Machine learning: a primer

Thu 18-01-2018
Machine learning extracts patterns from data without explicit instructions.

In this primer, we focus on essential ML principles— a modeling strategy to let the data speak for themselves, to the extent possible.

The benefits of ML arise from its use of a large number of tuning parameters or weights, which control the algorithm’s complexity and are estimated from the data using numerical optimization. Often ML algorithms are motivated by heuristics such as models of interacting neurons or natural evolution—even if the underlying mechanism of the biological system being studied is substantially different. The utility of ML algorithms is typically assessed empirically by how well extracted patterns generalize to new observations.

Martin Krzywinski @MKrzywinski mkweb.bcgsc.ca
Nature Methods Points of Significance column: Machine learning: a primer. (read)

We present a data scenario in which we fit to a model with 5 predictors using polynomials and show what to expect from ML when noise and sample size vary. We also demonstrate the consequences of excluding an important predictor or including a spurious one.

Bzdok, D., Krzywinski, M. & Altman, N. (2017) Points of Significance: Machine learning: a primer. Nature Methods 14:1119–1120.

...more about the Points of Significance column

Snowflake simulation

Tue 16-01-2018
Symmetric, beautiful and unique.

Just in time for the season, I've simulated a snow-pile of snowflakes based on the Gravner-Griffeath model.

Martin Krzywinski @MKrzywinski mkweb.bcgsc.ca
A few of the beautiful snowflakes generated by the Gravner-Griffeath model. (explore)

The work is described as a wintertime tale in In Silico Flurries: Computing a world of snow and co-authored with Jake Lever in the Scientific American SA Blog.

Gravner, J. & Griffeath, D. (2007) Modeling Snow Crystal Growth II: A mesoscopic lattice map with plausible dynamics.

Genes that make us sick

Wed 22-11-2017
Where disease hides in the genome.

My illustration of the location of genes in the human genome that are implicated in disease appears in The Objects that Power the Global Economy, a book by Quartz.

Martin Krzywinski @MKrzywinski mkweb.bcgsc.ca
The location of genes implicated in disease in the human genome, shown here as a spiral. (more...)

Ensemble methods: Bagging and random forests

Wed 22-11-2017
Many heads are better than one.

We introduce two common ensemble methods: bagging and random forests. Both of these methods repeat a statistical analysis on a bootstrap sample to improve the accuracy of the predictor. Our column shows these methods as applied to Classification and Regression Trees.

Martin Krzywinski @MKrzywinski mkweb.bcgsc.ca
Nature Methods Points of Significance column: Ensemble methods: Bagging and random forests. (read)

For example, we can sample the space of values more finely when using bagging with regression trees because each sample has potentially different boundaries at which the tree splits.

Random forests generate a large number of trees by not only generating bootstrap samples but also randomly choosing which predictor variables are considered at each split in the tree.

Krzywinski, M. & Altman, N. (2017) Points of Significance: Ensemble methods: bagging and random forests. Nature Methods 14:933–934.

Background reading

Krzywinski, M. & Altman, N. (2017) Points of Significance: Classification and regression trees. Nature Methods 14:757–758.

...more about the Points of Significance column