Reproducing the code

Yesterday, Friday, I got up, tea'd up, spruced up because I was Countryboy Going To Dublin for the 'annual' VIBE symposium. I, wearing my 2014 VIBE organiser hat,  had sent out a call to everyone to remember their name badge, so I had a chance of remembering who they were:

Goodo, looking forward. No registration means no name-badges so could everyone remember to bring one? from their last conference?  Just as a courtesy to confused old buffers and total newbies.
Thanks, Bob
The previous (2016) organiser added a note:
And/or a roll of blank stickers that people can just write their name onto. 
To which the the 2019 organiser replied.
Great; please bring one.
Clearly some people are more invested in the Science than interpersonal signalling. And IMO, a roll of stickers is A Bad Idea because they are not designed to stick to jumpers and it invites useless contributions as shown [R]: Bubba? Who he? Who his boss? Which is his poster? What group is he affiliated with? Should I get concerned when he brings out a banjo?

The presentations were wide-ranging and all interesting, even those that were whoof totally over my head. Molecular sequence analysis is amazingly ambitious, seeming to keep up the precipitous drop in the cost of generating the primary sequence data. My report of VIBE 2013 noted that analytical throughput increased 4000x-fold in a decade while the co$t per DNA base had fallen by a factor of 10 million in 20 years. In the early 00s, I was tasked with determining if certain classes of genes [those switched on in the cells of the heart, liver and lights] were clustered in the human genome, which had just been released into the public domain. I learned to program in Perl, robbed code from my fellow workers at the genomics, hacked it to do my bidding, learned how to display data graphically using the GD add-in and presented my progress regularly at the lab-meetings. It was a challenge, it was hard work, it was great fun and eventually my mates started robbing my GD code to use for their own projects. A small part of my work-flow was getting a program to work - my first binfo boss used to say "that's just a simple Fortran program, take you ten minutes" which was orders-of-magnitude true: it took maybe 90 minutes. But it took the rest of the day, and sometimes the rest of the week to be confident that the results spat out were true and meaningful. I won't say I was a programming god, but I could do the work even if I still don't feel comfortable with associative arrays.

The first talk on Friday was an over-my-head job: Research and Clinical Applications of Fully Reproducible Containerised Workflow Architecture. Dang, but that could mean anything at all to an old code-kludger like me. Even though I didn't recognise the names of any of the software tools - they having been developed over the last 7 or 8 years since I fell away from the coal-face of science and went back to teaching down the country - I followed with admiration the aspiration. Reproducibility is the key!  It's like using Latin names to talk to Hungarian ornithologists. IF you build a pipeline to integrate a wild variety of disparate but relevant data so that you can develop predictors or therapies or a cure for colon cancer THEN you really need it to be reproducible in Hungary or next year. 

I've been there, and I've been caught with my pants down. Maybe 15 years ago, I was working in St. Vincent's before it became SVUH the University Hospital. The colon cancer people had generated a noisy multivariate dataset and asked me to see if there was any signal there. I had a word with my pal Aedin [prev] before she went all ad Astra and to Harvard on us. She was the local ADE-4 guru and with her help I beat the colon cancer data into submission. I was skeptical about the results I'd generated; not because they were wrong, but because they explained so little of the variation in the data. I sent the analysis up the line, with my skeptical warning, and Team Colon were delighted because TOP2A, a known actor in the development of colon cancer, bobbed to the top of the roiling mess of data, you could see it there, barely visible, awash with statistical noise, but top of the heap of over expressed genes. The project wasn't written up immediately: the bloke who'd done the original analysis finished his rotation, his boss moved to a different hospital and her boss was too busy saving lives. I moved to a different institution too. Four years later, I got a call to redo the analysis because the paper was finally ready for launch and they were checking through their pipeline; and the graphics I'd generated were kinda ugly.

Well it was a freaking nightmare. The computer on which I'd done the original analysis had been left in StVs as obsolete. ADE-4 had been totally re-written, and was reluctant to install on the new computer. I couldn't remember how to get it to work and Aedin was 5,000km away in Harvard rather than up the road in UCD. I knuckled down and did what I was asked but that's when my hair-loss started. TOP2A was still tops, which was a relief, and the paper was published. If it was left to me, it would have been written off  rather than written up - but then my scientific publication record is woeful. Even with the same people and precisely the same data, reproducibility was a hard slog.

The small world connection here is that the presenter of the Containerised Workflow Architecture talk yesterday answers to the same boss as the boss-boss-boss I was serving in 2005 and 2009. In the intervening years, the quality and esp. the quantity of the data has gone up and up. By using containers in your workflow you can let collaborators and rivals replicate your findings and run precisely comparable analyses of their own. The science will be reproducible and we'll be nearer to eliminating colon cancer.  All good until a senior scientist at Friday's VIBE asked - what happens if there is a mistake in your code; will everyone be replicating that error?

Hmmmm! As Osric has it: A hit, a very palpable hit. Later in the day, I was talking to a handful of MSc students who'd come up from Galway for the meeting but missed the first talk. They are a useful mix of biologists and coders and I said that I'd been part of a similar group when their boss had moved from mathematical physics to biological sequence analysis and helped me with my Perl scripting. I hoped they were sharing their expertise and propping each other up. And I gave them an executive summary of the first (Containerised Workflow Architecture) talk of the day which they'd missed and emphasised the critical replicating that error question. Don't do what we used to do, I said, if you are all robbing each other's code there's a good chance that errors will propagate through the lab. If we use computers [and especially fancy pipelines of interconnected software] to find out something new, nifty and important then someone else should rewrite as much of the code as possible from scratch. There are a hundred different ways of computing any reasonably complicated analysis, if you have two independently generated solutions and they agree then your confidence in the result is more than doubled.  Ain't gonna happen though, everyone wants their own apple to polish.

Whoa Excel woes

The thing about software is that it is designed to make things easy - if it well designed. If things are easy to do then they are done often and possibly without sufficient thought. If you had to statistical analysis by hand, you'd make damned sure that you designed the experiment well so that the results of your analysis meant something.

If the statistical analysis is no more than 10 or 15 minutes making mystical passes with Excel then you'd be more inclined to have a punt at the experimental design fully prepared to do it again - properly - if the data turns out to be impossible to analyse. In my teens, I did quite a few statistical computations 'by hand' - working through columns of data calculating the square of each value, the sum of the squared values, the square of the sum of the values etc. All without a calculator because they hadn't been  invented in 1971. If you made an error it would propagate through the table and become a huge distorting wen on the face of the analysis. I must be the youngest person on the planet to use Charlier's Checks for internal consistency in a mighty table set up to calculate the variance of a dataset. Charlier's Checks required additional bookkeeping work, true, but they also gave you a solid confidence in the answer. Now you just lash the numbers into Excel and demand Descriptive Statistics to obtain not only the Variance, but the coefficients of kurtosis and skewness as well [example R]. Ten years later, when calculators were cheap and widely available, I helped construct one of the last hand-cranked commercial spreadsheets before Visicalc and Lotus1-2-3, the progenitors of Excel, swept such activity into the dustbin of history.

Excel is designed for businesses rather than science ans seems unduly driven by time and date. If you are a teeny bit careless the number 12 will become 12th Jan 1900 while 19,892 will be treated as my birthday 17th June 1954. Excel has a number of such helpful features which, if you're not careful,  will leave you with a red face if nit in tears. There was a paper in Genome Biology a couple of years ago by Ziemann, Eren and El-Osta, which exposed a widespread casualness about data-in-excel. Molecular biology and genome analysis are now capable of generating terabytes of data for half-nothing in money and time. Scientific papers would be unreadable if all the data were put in tables; if not unreadable, then at least many many pages long. The rules of engagement dictate that the raw data has to be made available, so that rivals, reviewers or referees can critically evaluate it. The data is, accordingly installed in Supplementary Tables somewhere on The Cloud. Z.E.E. wrote a bot which trawled through these Excel tables associated with a number of reputable Journals and looked for damn-foolishness, sloppy copy-editting, and culpable negligence. They found a great deal of errors like where the gene Septin2 aka Sept2 had been helpfully translated as 02-Sep or March1 [Membrane-Associated Ring Finger (C3HC4) 1, E3 Ubiquitin Protein Ligase] became 03-10-2006 because Excel decided these names looked like dates. More to point, none of the authors or editors of these tables had ever checked them through, let alone corrected the errors. Just by following this line of reasoning, they found errors in about 20 % of all the Suppl Tables they ran through their wringer. Do you have less confidence in the findings of the associated peer-reviewed papers if the data has such obvious feet-of-clay?

I found this paper through a link in MeFi pointing at a group calling itself The European Spreadsheet Risks Interest Group - EuSpRIG - (“yewsprig”) for short. They have compiled a long list of errors compounded by the casual use of Excel, which they call horror stories. Whatever about a few 'typos' in the supplementary tables [which clearly nobody has read] of some specialised genome analysis papers, the horror stories have consequence: loss of money, loss of credibility, loss of reputation and loss of jobs. It's the kind of trusting-the-software stupidity that had the HSE flagging patients as less than dead in 2017.

I was reading these horror stories with a mix of nerdiness, incredulity and complacence actually not me with the complacent. I never believed what I caused to be generated by writing software: I checked all the calculations by hand many times before I accepted the output. I did my best to push through some edgy data to see how robust my programs were. Just before leaving work and going home to surf the blogosphere looking for nerdnik copy, I had calculated the final marks and breakdown for the continuous assessment CA part of my human physiology course. As it is a small class and as I couldn't work out how to send this compendium through Blackboard, I emailed each student the relevant line from my summary Excel spreadsheet. It was a bit of a rush, so I didn't double-check the numbers - partly because the sky wasn't going to fall if I got things wrong. Indeed, part of the reason for sending out the marks was to solicit complaints and queries. I might have forgotten to credit a sick-note, for example. When I next checked my mail there was indeed a query/complaint that the individual marks from Oct, Dec, Jan & Feb were not consistent with the recently compiled executive summary. Ooops! There is a peculiar legacy issue with student names at The Institute. Some early kludge software had choked on apostrophes as Not Alphabetic so email aliases now come as O Reilly, O Donovan, O Beirne. But one of my students is not properly registered so I have her as O'Donovan. Excel sorts O Leary before O'Donovan and in a frenzy of cutting and pasting between Excel sheets I had managed to transpose O Leary's marks with O Donovan's. Only a Pink Face over this because it doesn't really matter but it is a lesson to be vigilant in future.

sleep early and often

The Wexford Science Café meets on the third Tuesday of every month . . . except when it doesn't. It has been lurching running for more than 4 years now and we've covered:
Stephen J Gould’s spandrels; Galena; Thomas Kuhn’s paradigms; Organic soil microbiome; Composting toilets; Water quality; Bacteria in food prep; Urination once again; Toxicity from botox to beer; Air-quality and asthma; Gravity waves; Neuroscience of torture; Greenland ice melt; Cider making; Zombies; Wolbachia and tropical diseases; Radon; Diabetes & Alzheimers; Marketing generic meds; Oroville dam crisis; March for Science; Allometry; Science book-swap; Erwin Schrodinger; Pheromones; Back-garden astronomy. 
All interesting and showing the reach of science and the collective interests of the WSC participants. I am quite religious about turning up even if it means driving nearly a hour from home to get there: It is one of the few social engagements I have outside of work and nuclear family so is important for my mental health.

Last Tuesday we heard that getting a good night's sleep is also vital for your mental health. One of our reg'lar participants in the WSC Happy Family is Mr Pill the Pharmacist, who has recently become a Daaaad and two years into the gig is still laboring under a sleep deficit as the wean frequently neglects to snooze the night through. If he was a Kiwi, he'd have an instructional video. What the child's father has noticed is that he can be unaccountably ratty at work - amazed at the stupidity of his customers; furious when things go wrong; narky with his colleagues. Then he put two and two together to realise that his anger descended when he'd endured a really wretched night.  He also floated a hypothesis, as yet unquantified, that most of the kids who present a script for ADHD medication are no way ADHD: they are just sleep deprived. When a family comes to the doctor's surgery with €60 and a troublesome restless teenager, it is impossible for everyone if they come out with only a suggestion that a good night's sleep is required . . . and why not lock up the youngster's phone before bedtime? They'd rather dose the trouble away . . . and then get really indignant when Young Jimmy scores a few Es at the weekend.

The hook on which our discussions about sleep were hoist was the book by Matthew Walker, British author of Why We Sleep and Prof of Neuroscience at UC Berkeley. It is Walker's contention that sleep hygiene is at the root of many woes: mental and physical health; success at work; success "ín bed"; that car crash; that slice of chocolate cake. The book was a surprise runaway best-seller in 2017, so must have rung a few bells (or jangled a few chains) with the reading public. Which is a rather diminished cohort because book-reading is soooo yesterday and pushed to the back of the closet as everyone embraces screens. I've written about the negative impact of the [blue] light used to make screens work. Dau.II pointed out that device content is designed to be seductive if not  addictive. What are we like? The evidence of the damaging effects of sleep deprivation is piling up while we're swiping just one more tweet long after after midnight.

The sleepy discussion at WSC led on the work of Annie Curtis and her Clock Lab, now at RCSI in Dublin. She is finding that disruptions to another cycle can have serious health consequences. Lots of things go round and round and up and down for us on a daily basis - core body temperature; sleep-wake; the bowels. Curtis has put together large datasets [like half a million middle-aged Brits] showing that other aspects of well-being and equanimity fluctuate on an annual basis. You're more likely to have a fatal heart-attack in the Winter; surrogate cardio markers also tip up as the days get shorter: we get fatter and our blood-pressure goes up significantly, for starters. And another large cohort study showed that clinical depression descends on many women in Winter. The underlying basis of these adverse events is the molecular clock which is ticking in us all: cranking up the immune system at certain times of the year (perhaps because that's when the pathogens are abroad??) with unexpected side-effects in other systems of the body. Those epidemiological studies are paralleled by lab-based studies in which the undulating concentration key molecular immune markers are tracked across time. What's not to love about someone who can use a quote from Wm. Shagsper as the title of one of her papers.

“Sleep that knits up the raveled sleave of care,

The death of each day's life, sore labor's bath,

Balm of hurt minds, great nature's second course,

Chief nourisher in life's feast.”

Night-caps off!
More Women in Science.

Iditarod goes Idatadog

We spent St Patrick's weekend at home; and, although Mohammed stayed away, 50 years of strong smart women sat down to Sunday dinner up the mountain: The Beloved in the Matriarch's chair, Dau.I and her SO, Dau.II, Dau.II's odd-mother, o-m's daughter. I won't bore you with the ages but the range was 50+ years. The youngest person present is beginning to run and the adult women present were full of cogent advice and encouragement. Someone cited Chris McDougall's Born to Run [bloboprev] because in the elite ultramarathon world, women are at least as good as the men. If you can run 100km, then your chromosome count [and consequent testosterone titre] is no handicap. I can run 100m, uphill if necessary, but I cannot speak at the end of it and I have neither the wind, nor the bottle, to run 1km let alone 100km. otoh, I did walk 700km in 1989 and 800km in 2004, so I have [blistered] skin in the game. Me I'd cite Feet in the Clouds: a tale of fell running and obsession by Richard Askwith as a source of inspiration for young runners.

My preferred charity is the RNLI because a) it does good but b) it allows people to be their very best = heroic. Running also allows people to be their best selves: against self; in competition but also in compassion and kindness. Marathons and a fortiori ultramarathons really sort the sheep from the goats. I've recently become a bit of a sofa-groupie for dog-sled racing like the Iditarod Race which is not a direct commemoration of the famous diphtheria antitoxin delivery to Nome, Alaska in 1925. The race happens every year since 1973 and takes at least 8 days to cover 1500 km through a winter wilderness by a sled and dog-team. This year the third person across the line was Jessie Rover. Susan Butcher won rhe race outright four times in the last century, while Libby Reynolds was first home to Nome in 1985. That was a dreadful weather year and Reynolds took 18 days to cover the course. This year a lot of attention has been on rookie Blair Braverman who finished 36th and six whole days behind the winners. While she was mushing on through the blizzards, her fan-base were raising $000$ of dollars  #uglydogs to support Alaskan schools. Commentary at MeFi.

BB was one of 17 (about a third) female entrants in 2019, including Jessie Royer. It has got to be harder graft to finish 36th after such an extended and calorie sapping journey than in the top three.. Anyway, our youngest visitor over the weekend is mad-about-the-dogs having acquired a puppy  a few weeks ago to walk and clean up after. It occurs to me that if Jamaica can field bobsled teams, both men and women, Ireland can surely furnish an entrant to the Iditarod in about 10 years time.

[You can stop reading right now unless you are fully nerded up]
Scoping out the results of this years gig, I noticed that one of the variables was Dogs In with numbers from N = 6 to N = 13. Hmmmm, I pondered, is the number of  Dogs In correlated with the time taken and/or is there an optimal number of dogs for getting over the line in the quickest time? That required cut&pasting the data from the results page in Excel and plotting Dogs In vs the time in seconds.

[Time in seconds = days*86,400 + hours*3600 + minutes*60 + seconds].

There are about 700,000 seconds in 8 days.

This picture is a special class of scatter-plot called a dot-plot where the range of values for each count of dogs is graphically displayed. This shows [me] that the range of race-times for a particular number of dogs is greater than any difference in the average time among dog-numbers. The display is far more important, and provides better insight into the distribution of the data than any ANOVA [analysis of variance] the standard statistical test for such data. The intuitive feel that here within group variance trumps mean values is actually borne out by the ANOVA. It takes about 1 million seconds or 11 and a half days to run the race regardless of the number of Dogs In. I'm not sure exactly what Dogs In signifies: are substitutions allowed during the race the way soccer players are ordered off the pitch by the coach and replaced by a player less puffed or less limpy?  Regardless of the details, I'll be using these data in Quant.Meths next year at The Institute to show the value of dot-plots.

Hemihomonyms

The other day, one of my project students was looking more competent than me [red face] in the tools of his/our trade. In too much of a hurry, I thought that Proboscia was essentially the same as Proboscidea. When I slowwwwwed down I twigged that one Proboscia is about 50μm on length while elephants are about about 100,000x longer trunk-to-tail. What a difference a '...de...' makes! The accident tinkled a distant bell that there were pairs of species which really did have the same name but were completely different.

A recent explanation of the rules and regs for the naming of parts in biology is The use and limits of scientific names in biological informatics by David Remsen Zookeys. 2016; (550): 207–223.  It's something that requires a certain amount of precision to prevent misunderstanding. If you want to communicate some measurements blue-tit [the bird in the picture] wings to your colleagues in Hungary then using a universal [dead] language prevents ambiguity: Cyanistes caeruleus is the only thing it could be I would be skeptical about using the Google - Magyar translate to kék cinege. Not a big problem in this example because the bird appears on both the EN and HU versions of wikipedia. But what about something more obscure: the White-bellied rice-rat Oryzomys albiventer isn't in hu.wikipedia.

One of the rules about taxonomy is naming priority. 25 years ago, I had a heart-in-boots moment on that front. We'd just published a sequence analysis paper on Aspergillus nidulans. and I thought I'd check the database to see if there were any new A. nidulans sequences which I could add to our tiny dataset. When I asked, ACNUC (the then cutting edge database interrogation software) came back with No Sequences Found. It was like arriving at the 13th Floor by elevator in an episode of The Twilight Zone. As Aspergillus nidulans is a widely studied standard genetic organism like lab mouse Mus musculus and Drosophila melanogaster, it wasn't just me who was hopping mad disconcerted. The naming people at GenBank had been following priority rules to rename the species as Emericella nidulans. Luckily they were pragmatists rather than pedants and, at the next full release of GenBank three months later, all my Aspergillus sequences were back on line.

The key element of nomenclature is to be unambiguous. If things are different they get different names. Careful analysis [bloboprev] of the biology and biogeography of giraffes Giraffa giraffa made experts believe that there were really 9 different species of long-necks in Africa and they all needed a separate species name. The names of animals are decided / approved by ICZN; a different body to the ICN namers of plants fungi and algae.
Partly from the inertia of the system; partly because the two entities are unambiguous different there are an least half a dozen hemihomonyms: species which have the same official Linnaean monnicker. Here [L] is a nice pair, both Agathis montana: LL is a braconid wasp widely distributed across Eurasia from Korea to UK. LR is a long-lived conifer from New Caledonia in the far east.  It is critically endangered from a perfect storm of bark-beetles delivering fungal disease, feral pigs, and habitat destruction from climate change and logging. That extinction will nicely tidy up the taxonomic conundrum, no? Here are the other hemihomonymous pairs acknowledged by wikipedia:

wha'?	Genus	species	whaaa'?
wasp	Agathis	montana	conifer
starfish	Asterina	gibbosa	fungus
marigold	Baileya	australis	moth
venomous fish	Centropogon	australis	bell-flower
another fish	Orestias	elegans	orchis
nudibranch	Tritonia	pallida	iris

That's only the top of the iceberg, however, Алексей Шипунов from Russia has compiled a mighty database of hemihomonyms - including 12 examples where a Genus is triplicated in bacteria, plants and animals:  Catenococcus  Gordonia  Kingella  Lawsonia  Leptonema  Microcyclus  Moorella  Morganella  Rhodococcus  Rothia  Spirulina  Stenocybe.

Because these worthy taxonomy-mavens are aware of illiteracy, dyslexia and Latinophobia among those who need to know the Names of Life, there is another class of wrong-wrong-almost-right called parahomonyms which are at best discouraged if not absolutely verboten.  Thus Astrostemma Benth. 1880 has been rejected in favour of Absolmia Kuntze; because it was too much like Asterostemma Decne. 1838 another genus of plants in the same family. To crowd confusion on ambiguity Asterostemma depressa is a fossil mammal from Argentina.

A couple of years ago I was laying out a very similar problem with the naming of drugs: where errors are more likely to kills people than mistakes

Kosherduck

Kosher is a set of dietary laws which Jewish people take with a pinch of salt. Actually rather a lot of salt because cuts of meat are often ladled with the stuff to draw out the last traces of visible blood. I've worried about whether rabbits are kosher - they are not kosher because they chew the poo rather than chew the cud to max out the nutritional value of what they eat. Camels are not kosher on the other side of the definition - they indeed chew the cud but the hoof is not split.  Whaaa'? Looks pretty goddamned split to me [R]. Note to self: do not google "camel toe".

On the fish front, the definition is also two-fold: if the cr'ature has both scales and fins it is good to eat. Weirdly if it has scales then it is assumed to have fins. This excludes eels, sturgeon (and caviar), catfish, sword-fish . . . and of course a whole rattle of not-fish like shrimp, lobster, crab, mussels, scallops, whales and octopus. I think that one of the delights of kosher for Jews is that it encourages pedantic arguments about whether a kosher fish found in the belly of not-kosher fish is kosher - it is.

The Lord wasn't able to give a neat paired attributes {scale&fins | cud&hooves} definition of permitted birds. Instead the Torah lists -in Hebrew, natch  - 24 forbidden birds - none of which had the unambiguous Linnaean Latin binomer. And even if they had a clear label, would an averagely educated Jewish housekeeper recognise Otus scops or שעיר, שעיר מצו or Zwergohreule if it jumped up and bit him? That's rather inclusive: there being 10,000 different species of bird, only 2 dozen of which are suspect. Ambiguity and identity problems were only a challenge to the scholars of the yeshiva: nothing better than poring over an old manuscript to read the opinions of a long-dead scholastic rabbi from Lvov. The Torah is the word of god (equivalent to the Christian Olde Testament) the subsequent notes, criticism, arguments and counter arguments form the Mishnah, Gemara and Talmud.

If the forbidden birds have anything in common, it might be that they are hawks, eagles, owls and other raptors: birds that eat other vertebrates. The scholars of the Mishnah came to agree on three other 'helpful' criteria

• having an extra toe
• having a crop [a muscular pocket for grinding seeds etc.before the contents osdelivered to the stomach - some birds swallow small stones to help the grist-mill]
• having a gizzard that can be peeled

all these characteristics are shared by chickens. You have to suspect that the scholars were stacking the discriminant function so that, whatever got spat out as not kosher, chicken soup and schmalz was allowable.
That diligent scholarship in the medieval ghettos of Lviv [priv] and Warsawa to determine what a good jew can eat without damageto the immortal soul has been recently emulated on Tumblr.  Here is a list of the (alas all too few = 77/800+) Pokemon which are kosher. Like with the Mishnah dealing with real animals, with clear rules most of the decisions are easy. It is the edge-cases which challenge the scholar; the grey areas; the equivocal; the noisy; the things that are not internally consistent w.r.t. to the rules of the kosher game:

• Stantler: kosher, in my opinion. the split hooves are only visible from some angles and in some sprites. Consult your local Rabbi. 
• Psyduck and Golduck: perhaps controversial, but i am going directly AGAINST the Kashrut Laws in Pokémon guide by declaring these two kosher. yes, they’re water types, and could be arguably treated like fish, but look at them; they’re ducks 
• Cherubi and Cherrim are kosher. if you make them into wine, a Rabbi must be present. 
• Skiddo and Gogoat are doubly kosher, as both legitimate goats with split hooves AND plants.

The names mean nothing to me; the SOs of Dau.I and Dau.II both know how to spell ポケットモンスター though.

sh sh sh Sherborne

Sax six *** sox sux

• Sherborne Station watch a big chap fill out with Big Boy Breakfast. 
• Would be better at 3.5 minutes rather than 35m
• Sherborne is where me mother goes shopping on Thursdays
• Spectre start Craig is Bond
• Sean as Sharpe o'er the hills and far away again?
• Scything, how to
• Sally le Page goes bananas
• Her Shed Science channel on baking and gluten
• Sourdough - don't give up on your dreams
• Don't give up - Pete & Kate
• Don't give in - Snow Patrol
• Stones - lithographic prints
• Sanctus - Messe de Sainte Cecile; Gounod
• Soulé - Love no More [never 'eard of her tho a Dublin sensation]
• Steppes of Central Asia В средней Азии - dances by Borodin.
• Sofa geometry
• Sedaris, David on seat-giving: strangely reminding me of the folks who are agonised to like the music of Michael Jackson after he was condemned for being a Bad Man
• Sintaklaas packs kids into a sack and sends them to Spain: Six to Eight black men
• Swear or affirm - why Sinn Fein don't sit in the UK Parliament. Data! Who affirms? Who swears and on what book [plenty choice]; and in which language?